diff --git "a/nohup.out" "b/nohup.out" --- "a/nohup.out" +++ "b/nohup.out" @@ -54963,3 +54963,11207 @@ huggingface/tokenizers: The current process just got forked, after parallelism h To disable this warning, you can either: - Avoid using `tokenizers` before the fork if possible - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) + + Training...: 11% 463/4393 [45:57<182:26:10, 167.12s/it]/home/sanchitgandhi/hf/lib/python3.8/site-packages/flax/training/common_utils.py:25: FutureWarning: jax.tree_map is deprecated, and will be removed in a future release. Use jax.tree_util.tree_map instead. + return jax.tree_map( + + Training...: 11% 464/4393 [46:03<129:46:59, 118.92s/it] + Training...: 11% 465/4393 [46:09<92:43:14, 84.98s/it]  + Training...: 11% 466/4393 [46:15<66:45:49, 61.20s/it] + Training...: 11% 467/4393 [46:21<48:35:59, 44.56s/it] + Training...: 11% 468/4393 [46:26<35:51:00, 32.88s/it] + Training...: 11% 469/4393 [46:32<27:00:34, 24.78s/it] + Training...: 11% 470/4393 [46:38<20:42:41, 19.01s/it] + Training...: 11% 471/4393 [46:43<16:20:25, 15.00s/it] + Training...: 11% 472/4393 [46:49<13:12:59, 12.13s/it] + Training...: 11% 473/4393 [46:54<11:04:21, 10.17s/it] + Training...: 11% 474/4393 [47:00<9:30:29, 8.73s/it]  + Training...: 11% 475/4393 [47:05<8:23:44, 7.71s/it] + Training...: 11% 476/4393 [47:10<7:36:44, 7.00s/it] + Training...: 11% 477/4393 [47:16<7:03:54, 6.49s/it] + Training...: 11% 478/4393 [47:21<6:41:06, 6.15s/it] + Training...: 11% 479/4393 [47:26<6:24:57, 5.90s/it] + Training...: 11% 480/4393 [47:31<6:07:51, 5.64s/it] + Training...: 11% 481/4393 [47:36<5:55:16, 5.45s/it] + Training...: 11% 482/4393 [47:41<5:43:46, 5.27s/it] + Training...: 11% 483/4393 [47:46<5:35:46, 5.15s/it] + Training...: 11% 484/4393 [47:51<5:26:20, 5.01s/it] + Training...: 11% 485/4393 [47:55<5:18:16, 4.89s/it] + Training...: 11% 486/4393 [48:00<5:08:34, 4.74s/it] + Training...: 11% 487/4393 [48:04<5:01:45, 4.64s/it]/home/sanchitgandhi/hf/lib/python3.8/site-packages/flax/jax_utils.py:61: FutureWarning: jax.tree_map is deprecated, and will be removed in a future release. Use jax.tree_util.tree_map instead. + return jax.tree_map(lambda x: x[0], tree) +run_flax_speech_recognition_seq2seq.py:336: FutureWarning: jax.tree_map is deprecated, and will be removed in a future release. Use jax.tree_util.tree_map instead. + return jax.tree_map(lambda x: x.astype(jnp.float32) if x.dtype == jnp.bfloat16 else x, t) + +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:25:09<17:31:17, 21025.84s/it] + Training...: 11% 487/4393 [48:09<5:01:45, 4.64s/it] + Training...: 11% 488/4393 [48:09<5:04:13, 4.67s/it] + Training...: 11% 489/4393 [48:13<4:50:03, 4.46s/it] + Training...: 11% 490/4393 [48:17<4:36:33, 4.25s/it] + Training...: 11% 491/4393 [48:20<4:21:00, 4.01s/it] + Training...: 11% 492/4393 [48:23<4:05:23, 3.77s/it] + Training...: 11% 493/4393 [48:26<3:47:00, 3.49s/it] + Training...: 11% 494/4393 [48:29<3:31:36, 3.26s/it] + Training...: 11% 495/4393 [48:31<3:14:04, 2.99s/it] + Training...: 11% 496/4393 [48:33<2:57:13, 2.73s/it] + Training...: 11% 497/4393 [48:35<2:40:39, 2.47s/it] + Training...: 11% 498/4393 [48:37<2:24:23, 2.22s/it] + Training...: 11% 499/4393 [48:38<2:06:49, 1.95s/it] + Training...: 11% 500/4393 [48:39<1:49:29, 1.69s/it] + Training...: 11% 501/4393 [48:46<3:22:17, 3.12s/it] + Training...: 11% 502/4393 [48:52<4:25:48, 4.10s/it] + Training...: 11% 503/4393 [48:58<5:04:00, 4.69s/it] + Training...: 11% 504/4393 [49:04<5:28:53, 5.07s/it] + Training...: 11% 505/4393 [49:10<5:50:40, 5.41s/it] + Training...: 12% 506/4393 [49:17<6:02:52, 5.60s/it] + Training...: 12% 507/4393 [49:23<6:11:02, 5.73s/it] + Training...: 12% 508/4393 [49:28<6:14:55, 5.79s/it] + Training...: 12% 509/4393 [49:34<6:17:17, 5.83s/it] + Training...: 12% 510/4393 [49:40<6:15:56, 5.81s/it] + Training...: 12% 511/4393 [49:46<6:15:08, 5.80s/it] + Training...: 12% 512/4393 [49:52<6:14:19, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:26:58<17:31:17, 21025.84s/it] + Training...: 12% 512/4393 [49:58<6:14:19, 5.79s/it] + Training...: 12% 513/4393 [49:58<6:29:55, 6.03s/it] + Training...: 12% 514/4393 [50:04<6:24:29, 5.95s/it] + Training...: 12% 515/4393 [50:10<6:19:52, 5.88s/it] + Training...: 12% 516/4393 [50:15<6:14:05, 5.79s/it] + Training...: 12% 517/4393 [50:21<6:10:16, 5.73s/it] + Training...: 12% 518/4393 [50:26<6:06:22, 5.67s/it] + Training...: 12% 519/4393 [50:32<6:04:00, 5.64s/it] + Training...: 12% 520/4393 [50:37<5:59:38, 5.57s/it] + Training...: 12% 521/4393 [50:43<5:56:53, 5.53s/it] + Training...: 12% 522/4393 [50:48<5:53:20, 5.48s/it] + Training...: 12% 523/4393 [50:54<5:53:03, 5.47s/it] + Training...: 12% 524/4393 [50:59<5:50:07, 5.43s/it] + Training...: 12% 525/4393 [51:04<5:46:41, 5.38s/it] + Training...: 12% 526/4393 [51:10<5:44:28, 5.34s/it] + Training...: 12% 527/4393 [51:15<5:45:48, 5.37s/it] + Training...: 12% 528/4393 [51:20<5:42:23, 5.32s/it] + Training...: 12% 529/4393 [51:25<5:38:59, 5.26s/it] + Training...: 12% 530/4393 [51:30<5:34:46, 5.20s/it] + Training...: 12% 531/4393 [51:35<5:31:39, 5.15s/it] + Training...: 12% 532/4393 [51:40<5:25:47, 5.06s/it] + Training...: 12% 533/4393 [51:45<5:20:37, 4.98s/it] + Training...: 12% 534/4393 [51:50<5:15:13, 4.90s/it] + Training...: 12% 535/4393 [51:54<5:08:55, 4.80s/it] + Training...: 12% 536/4393 [51:59<5:01:17, 4.69s/it] + Training...: 12% 537/4393 [52:03<4:52:55, 4.56s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:29:08<17:31:17, 21025.84s/it] + Training...: 12% 537/4393 [52:08<4:52:55, 4.56s/it] + Training...: 12% 538/4393 [52:08<4:54:17, 4.58s/it] + Training...: 12% 539/4393 [52:12<4:40:41, 4.37s/it] + Training...: 12% 540/4393 [52:15<4:26:28, 4.15s/it] + Training...: 12% 541/4393 [52:19<4:11:09, 3.91s/it] + Training...: 12% 542/4393 [52:22<3:55:29, 3.67s/it] + Training...: 12% 543/4393 [52:24<3:38:48, 3.41s/it] + Training...: 12% 544/4393 [52:27<3:22:16, 3.15s/it] + Training...: 12% 545/4393 [52:29<3:05:07, 2.89s/it] + Training...: 12% 546/4393 [52:31<2:48:32, 2.63s/it] + Training...: 12% 547/4393 [52:33<2:31:16, 2.36s/it] + Training...: 12% 548/4393 [52:35<2:14:57, 2.11s/it] + Training...: 12% 549/4393 [52:36<1:59:25, 1.86s/it] + Training...: 13% 550/4393 [52:37<1:43:24, 1.61s/it] + Training...: 13% 551/4393 [52:43<3:13:15, 3.02s/it] + Training...: 13% 552/4393 [52:49<4:14:50, 3.98s/it] + Training...: 13% 553/4393 [52:56<4:57:43, 4.65s/it] + Training...: 13% 554/4393 [53:02<5:25:09, 5.08s/it] + Training...: 13% 555/4393 [53:08<5:42:05, 5.35s/it] + Training...: 13% 556/4393 [53:14<5:53:07, 5.52s/it] + Training...: 13% 557/4393 [53:19<6:00:30, 5.64s/it] + Training...: 13% 558/4393 [53:25<6:04:37, 5.70s/it] + Training...: 13% 559/4393 [53:31<6:13:24, 5.84s/it] + Training...: 13% 560/4393 [53:37<6:13:37, 5.85s/it] + Training...: 13% 561/4393 [53:43<6:12:29, 5.83s/it] + Training...: 13% 562/4393 [53:49<6:10:53, 5.81s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:30:55<17:31:17, 21025.84s/it] + Training...: 13% 562/4393 [53:56<6:10:53, 5.81s/it] + Training...: 13% 563/4393 [53:56<6:25:58, 6.05s/it] + Training...: 13% 564/4393 [54:01<6:20:27, 5.96s/it] + Training...: 13% 565/4393 [54:07<6:16:02, 5.89s/it] + Training...: 13% 566/4393 [54:13<6:12:14, 5.84s/it] + Training...: 13% 567/4393 [54:18<6:08:36, 5.78s/it] + Training...: 13% 568/4393 [54:24<6:05:52, 5.74s/it] + Training...: 13% 569/4393 [54:30<6:03:17, 5.70s/it] + Training...: 13% 570/4393 [54:35<5:59:02, 5.63s/it] + Training...: 13% 571/4393 [54:41<5:58:59, 5.64s/it] + Training...: 13% 572/4393 [54:46<6:00:42, 5.66s/it] + Training...: 13% 573/4393 [54:52<5:58:49, 5.64s/it] + Training...: 13% 574/4393 [54:57<5:55:05, 5.58s/it] + Training...: 13% 575/4393 [55:03<5:53:11, 5.55s/it] + Training...: 13% 576/4393 [55:08<5:51:28, 5.52s/it] + Training...: 13% 577/4393 [55:14<5:49:40, 5.50s/it] + Training...: 13% 578/4393 [55:19<5:44:52, 5.42s/it] + Training...: 13% 579/4393 [55:24<5:42:17, 5.38s/it] + Training...: 13% 580/4393 [55:30<5:37:45, 5.31s/it] + Training...: 13% 581/4393 [55:35<5:33:01, 5.24s/it] + Training...: 13% 582/4393 [55:40<5:28:08, 5.17s/it] + Training...: 13% 583/4393 [55:45<5:25:58, 5.13s/it] + Training...: 13% 584/4393 [55:50<5:21:15, 5.06s/it] + Training...: 13% 585/4393 [55:54<5:14:58, 4.96s/it] + Training...: 13% 586/4393 [55:59<5:09:32, 4.88s/it] + Training...: 13% 587/4393 [56:03<5:02:11, 4.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:33:08<17:31:17, 21025.84s/it] + Training...: 13% 587/4393 [56:08<5:02:11, 4.76s/it] + Training...: 13% 588/4393 [56:08<5:04:12, 4.80s/it] + Training...: 13% 589/4393 [56:12<4:48:04, 4.54s/it] + Training...: 13% 590/4393 [56:16<4:33:15, 4.31s/it] + Training...: 13% 591/4393 [56:20<4:19:43, 4.10s/it] + Training...: 13% 592/4393 [56:23<4:04:38, 3.86s/it] + Training...: 13% 593/4393 [56:26<3:48:28, 3.61s/it] + Training...: 14% 594/4393 [56:29<3:32:30, 3.36s/it] + Training...: 14% 595/4393 [56:31<3:14:50, 3.08s/it] + Training...: 14% 596/4393 [56:33<2:57:42, 2.81s/it] + Training...: 14% 597/4393 [56:35<2:41:04, 2.55s/it] + Training...: 14% 598/4393 [56:37<2:24:24, 2.28s/it] + Training...: 14% 599/4393 [56:38<2:07:23, 2.01s/it] + Training...: 14% 600/4393 [56:39<1:49:29, 1.73s/it] + Training...: 14% 601/4393 [56:46<3:13:48, 3.07s/it] + Training...: 14% 602/4393 [56:52<4:14:59, 4.04s/it] + Training...: 14% 603/4393 [56:58<4:54:02, 4.66s/it] + Training...: 14% 604/4393 [57:04<5:19:20, 5.06s/it] + Training...: 14% 605/4393 [57:10<5:37:04, 5.34s/it] + Training...: 14% 606/4393 [57:16<5:48:51, 5.53s/it] + Training...: 14% 607/4393 [57:22<5:56:34, 5.65s/it] + Training...: 14% 608/4393 [57:28<6:00:53, 5.72s/it] + Training...: 14% 609/4393 [57:34<6:03:39, 5.77s/it] + Training...: 14% 610/4393 [57:39<6:04:21, 5.78s/it] + Training...: 14% 611/4393 [57:45<6:07:22, 5.83s/it] + Training...: 14% 612/4393 [57:51<6:08:59, 5.86s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:34:58<17:31:17, 21025.84s/it] + Training...: 14% 612/4393 [57:58<6:08:59, 5.86s/it] + Training...: 14% 613/4393 [57:58<6:21:44, 6.06s/it] + Training...: 14% 614/4393 [58:04<6:13:43, 5.93s/it] + Training...: 14% 615/4393 [58:09<6:08:33, 5.85s/it] + Training...: 14% 616/4393 [58:15<6:04:29, 5.79s/it] + Training...: 14% 617/4393 [58:21<6:03:36, 5.78s/it] + Training...: 14% 618/4393 [58:26<5:58:35, 5.70s/it] + Training...: 14% 619/4393 [58:32<5:54:31, 5.64s/it] + Training...: 14% 620/4393 [58:37<5:49:46, 5.56s/it] + Training...: 14% 621/4393 [58:42<5:46:26, 5.51s/it] + Training...: 14% 622/4393 [58:48<5:43:49, 5.47s/it] + Training...: 14% 623/4393 [58:53<5:42:06, 5.44s/it] + Training...: 14% 624/4393 [58:58<5:39:36, 5.41s/it] + Training...: 14% 625/4393 [59:04<5:36:00, 5.35s/it] + Training...: 14% 626/4393 [59:09<5:32:23, 5.29s/it] + Training...: 14% 627/4393 [59:14<5:29:11, 5.24s/it] + Training...: 14% 628/4393 [59:19<5:25:47, 5.19s/it] + Training...: 14% 629/4393 [59:24<5:22:31, 5.14s/it] + Training...: 14% 630/4393 [59:29<5:18:01, 5.07s/it] + Training...: 14% 631/4393 [59:34<5:14:25, 5.01s/it] + Training...: 14% 632/4393 [59:39<5:11:11, 4.96s/it] + Training...: 14% 633/4393 [59:43<5:05:57, 4.88s/it] + Training...: 14% 634/4393 [59:48<4:59:53, 4.79s/it] + Training...: 14% 635/4393 [59:52<4:54:05, 4.70s/it] + Training...: 14% 636/4393 [59:57<4:45:38, 4.56s/it] + Training...: 15% 637/4393 [1:00:01<4:38:46, 4.45s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:37:05<17:31:17, 21025.84s/it] + Training...: 15% 637/4393 [1:00:06<4:38:46, 4.45s/it] + Training...: 15% 638/4393 [1:00:06<4:42:08, 4.51s/it] + Training...: 15% 639/4393 [1:00:09<4:30:22, 4.32s/it] + Training...: 15% 640/4393 [1:00:13<4:19:27, 4.15s/it] + Training...: 15% 641/4393 [1:00:17<4:06:02, 3.93s/it] + Training...: 15% 642/4393 [1:00:20<3:52:09, 3.71s/it] + Training...: 15% 643/4393 [1:00:23<3:36:31, 3.46s/it] + Training...: 15% 644/4393 [1:00:25<3:21:35, 3.23s/it] + Training...: 15% 645/4393 [1:00:28<3:05:31, 2.97s/it] + Training...: 15% 646/4393 [1:00:30<2:49:43, 2.72s/it] + Training...: 15% 647/4393 [1:00:32<2:32:52, 2.45s/it] + Training...: 15% 648/4393 [1:00:33<2:16:49, 2.19s/it] + Training...: 15% 649/4393 [1:00:35<2:00:31, 1.93s/it] + Training...: 15% 650/4393 [1:00:36<1:43:52, 1.67s/it] + Training...: 15% 651/4393 [1:00:42<3:10:43, 3.06s/it] + Training...: 15% 652/4393 [1:00:48<4:12:23, 4.05s/it] + Training...: 15% 653/4393 [1:00:54<4:50:52, 4.67s/it] + Training...: 15% 654/4393 [1:01:00<5:17:25, 5.09s/it] + Training...: 15% 655/4393 [1:01:06<5:33:26, 5.35s/it] + Training...: 15% 656/4393 [1:01:12<5:43:51, 5.52s/it] + Training...: 15% 657/4393 [1:01:18<5:50:50, 5.63s/it] + Training...: 15% 658/4393 [1:01:24<5:55:24, 5.71s/it] + Training...: 15% 659/4393 [1:01:30<5:59:47, 5.78s/it] + Training...: 15% 660/4393 [1:01:36<6:01:16, 5.81s/it] + Training...: 15% 661/4393 [1:01:42<6:01:06, 5.81s/it] + Training...: 15% 662/4393 [1:01:48<6:00:16, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:38:54<17:31:17, 21025.84s/it] + Training...: 15% 662/4393 [1:01:54<6:00:16, 5.79s/it] + Training...: 15% 663/4393 [1:01:54<6:14:38, 6.03s/it] + Training...: 15% 664/4393 [1:02:00<6:08:54, 5.94s/it] + Training...: 15% 665/4393 [1:02:05<6:03:26, 5.85s/it] + Training...: 15% 666/4393 [1:02:11<6:00:23, 5.80s/it] + Training...: 15% 667/4393 [1:02:17<5:59:37, 5.79s/it] + Training...: 15% 668/4393 [1:02:22<5:55:44, 5.73s/it] + Training...: 15% 669/4393 [1:02:28<5:50:39, 5.65s/it] + Training...: 15% 670/4393 [1:02:33<5:47:28, 5.60s/it] + Training...: 15% 671/4393 [1:02:39<5:44:13, 5.55s/it] + Training...: 15% 672/4393 [1:02:44<5:43:58, 5.55s/it] + Training...: 15% 673/4393 [1:02:50<5:42:38, 5.53s/it] + Training...: 15% 674/4393 [1:02:55<5:38:12, 5.46s/it] + Training...: 15% 675/4393 [1:03:01<5:35:54, 5.42s/it] + Training...: 15% 676/4393 [1:03:06<5:31:53, 5.36s/it] + Training...: 15% 677/4393 [1:03:11<5:28:59, 5.31s/it] + Training...: 15% 678/4393 [1:03:16<5:25:58, 5.26s/it] + Training...: 15% 679/4393 [1:03:21<5:24:35, 5.24s/it] + Training...: 15% 680/4393 [1:03:26<5:20:37, 5.18s/it] + Training...: 16% 681/4393 [1:03:31<5:15:53, 5.11s/it] + Training...: 16% 682/4393 [1:03:36<5:10:39, 5.02s/it] + Training...: 16% 683/4393 [1:03:41<5:05:47, 4.95s/it] + Training...: 16% 684/4393 [1:03:46<5:00:54, 4.87s/it] + Training...: 16% 685/4393 [1:03:50<4:55:11, 4.78s/it] + Training...: 16% 686/4393 [1:03:55<4:48:23, 4.67s/it] + Training...: 16% 687/4393 [1:03:59<4:41:18, 4.55s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:41:03<17:31:17, 21025.84s/it] + Training...: 16% 687/4393 [1:04:04<4:41:18, 4.55s/it] + Training...: 16% 688/4393 [1:04:04<4:44:52, 4.61s/it] + Training...: 16% 689/4393 [1:04:07<4:30:52, 4.39s/it] + Training...: 16% 690/4393 [1:04:11<4:17:33, 4.17s/it] + Training...: 16% 691/4393 [1:04:15<4:03:32, 3.95s/it] + Training...: 16% 692/4393 [1:04:18<3:49:16, 3.72s/it] + Training...: 16% 693/4393 [1:04:21<3:34:22, 3.48s/it] + Training...: 16% 694/4393 [1:04:23<3:18:56, 3.23s/it] + Training...: 16% 695/4393 [1:04:26<3:03:06, 2.97s/it] + Training...: 16% 696/4393 [1:04:28<2:46:29, 2.70s/it] + Training...: 16% 697/4393 [1:04:30<2:30:32, 2.44s/it] + Training...: 16% 698/4393 [1:04:31<2:15:10, 2.20s/it] + Training...: 16% 699/4393 [1:04:32<1:58:49, 1.93s/it] + Training...: 16% 700/4393 [1:04:33<1:41:48, 1.65s/it] + Training...: 16% 701/4393 [1:04:40<3:06:40, 3.03s/it] + Training...: 16% 702/4393 [1:04:46<4:04:53, 3.98s/it] + Training...: 16% 703/4393 [1:04:52<4:44:19, 4.62s/it] + Training...: 16% 704/4393 [1:04:58<5:10:55, 5.06s/it] + Training...: 16% 705/4393 [1:05:04<5:30:45, 5.38s/it] + Training...: 16% 706/4393 [1:05:10<5:41:10, 5.55s/it] + Training...: 16% 707/4393 [1:05:16<5:46:57, 5.65s/it] + Training...: 16% 708/4393 [1:05:22<5:49:50, 5.70s/it] + Training...: 16% 709/4393 [1:05:28<5:52:09, 5.74s/it] + Training...: 16% 710/4393 [1:05:33<5:52:07, 5.74s/it] + Training...: 16% 711/4393 [1:05:39<5:53:05, 5.75s/it] + Training...: 16% 712/4393 [1:05:45<5:52:35, 5.75s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:42:51<17:31:17, 21025.84s/it] + Training...: 16% 712/4393 [1:05:51<5:52:35, 5.75s/it] + Training...: 16% 713/4393 [1:05:51<6:06:41, 5.98s/it] + Training...: 16% 714/4393 [1:05:57<6:01:49, 5.90s/it] + Training...: 16% 715/4393 [1:06:03<5:58:43, 5.85s/it] + Training...: 16% 716/4393 [1:06:09<5:54:06, 5.78s/it] + Training...: 16% 717/4393 [1:06:14<5:53:05, 5.76s/it] + Training...: 16% 718/4393 [1:06:20<5:49:40, 5.71s/it] + Training...: 16% 719/4393 [1:06:25<5:46:22, 5.66s/it] + Training...: 16% 720/4393 [1:06:31<5:45:38, 5.65s/it] + Training...: 16% 721/4393 [1:06:36<5:41:54, 5.59s/it] + Training...: 16% 722/4393 [1:06:42<5:36:55, 5.51s/it] + Training...: 16% 723/4393 [1:06:47<5:33:06, 5.45s/it] + Training...: 16% 724/4393 [1:06:52<5:29:12, 5.38s/it] + Training...: 17% 725/4393 [1:06:58<5:26:36, 5.34s/it] + Training...: 17% 726/4393 [1:07:03<5:23:36, 5.29s/it] + Training...: 17% 727/4393 [1:07:08<5:20:06, 5.24s/it] + Training...: 17% 728/4393 [1:07:13<5:17:01, 5.19s/it] + Training...: 17% 729/4393 [1:07:18<5:14:12, 5.15s/it] + Training...: 17% 730/4393 [1:07:23<5:09:22, 5.07s/it] + Training...: 17% 731/4393 [1:07:28<5:05:57, 5.01s/it] + Training...: 17% 732/4393 [1:07:33<5:01:42, 4.94s/it] + Training...: 17% 733/4393 [1:07:37<4:58:30, 4.89s/it] + Training...: 17% 734/4393 [1:07:42<4:54:12, 4.82s/it] + Training...: 17% 735/4393 [1:07:47<4:51:23, 4.78s/it] + Training...: 17% 736/4393 [1:07:51<4:42:42, 4.64s/it] + Training...: 17% 737/4393 [1:07:55<4:34:40, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:45:00<17:31:17, 21025.84s/it] + Training...: 17% 737/4393 [1:08:00<4:34:40, 4.51s/it] + Training...: 17% 738/4393 [1:08:00<4:36:22, 4.54s/it] + Training...: 17% 739/4393 [1:08:04<4:22:23, 4.31s/it] + Training...: 17% 740/4393 [1:08:07<4:08:00, 4.07s/it] + Training...: 17% 741/4393 [1:08:10<3:54:13, 3.85s/it] + Training...: 17% 742/4393 [1:08:13<3:39:16, 3.60s/it] + Training...: 17% 743/4393 [1:08:16<3:24:06, 3.36s/it] + Training...: 17% 744/4393 [1:08:19<3:07:14, 3.08s/it] + Training...: 17% 745/4393 [1:08:21<2:51:14, 2.82s/it] + Training...: 17% 746/4393 [1:08:23<2:35:48, 2.56s/it] + Training...: 17% 747/4393 [1:08:25<2:21:14, 2.32s/it] + Training...: 17% 748/4393 [1:08:26<2:07:26, 2.10s/it] + Training...: 17% 749/4393 [1:08:27<1:52:50, 1.86s/it] + Training...: 17% 750/4393 [1:08:28<1:37:02, 1.60s/it] + Training...: 17% 751/4393 [1:08:35<3:01:52, 3.00s/it] + Training...: 17% 752/4393 [1:08:41<4:01:26, 3.98s/it] + Training...: 17% 753/4393 [1:08:47<4:40:23, 4.62s/it] + Training...: 17% 754/4393 [1:08:53<5:05:41, 5.04s/it] + Training...: 17% 755/4393 [1:08:59<5:23:56, 5.34s/it] + Training...: 17% 756/4393 [1:09:05<5:39:37, 5.60s/it] + Training...: 17% 757/4393 [1:09:11<5:46:51, 5.72s/it] + Training...: 17% 758/4393 [1:09:17<5:48:29, 5.75s/it] + Training...: 17% 759/4393 [1:09:23<5:49:45, 5.77s/it] + Training...: 17% 760/4393 [1:09:29<5:49:52, 5.78s/it] + Training...: 17% 761/4393 [1:09:35<5:48:44, 5.76s/it] + Training...: 17% 762/4393 [1:09:40<5:49:39, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:46:47<17:31:17, 21025.84s/it] + Training...: 17% 762/4393 [1:09:47<5:49:39, 5.78s/it] + Training...: 17% 763/4393 [1:09:47<6:03:34, 6.01s/it] + Training...: 17% 764/4393 [1:09:52<5:56:17, 5.89s/it] + Training...: 17% 765/4393 [1:09:58<5:52:01, 5.82s/it] + Training...: 17% 766/4393 [1:10:04<5:49:00, 5.77s/it] + Training...: 17% 767/4393 [1:10:10<5:47:42, 5.75s/it] + Training...: 17% 768/4393 [1:10:15<5:43:43, 5.69s/it] + Training...: 18% 769/4393 [1:10:21<5:39:44, 5.62s/it] + Training...: 18% 770/4393 [1:10:26<5:37:26, 5.59s/it] + Training...: 18% 771/4393 [1:10:32<5:35:15, 5.55s/it] + Training...: 18% 772/4393 [1:10:37<5:30:33, 5.48s/it] + Training...: 18% 773/4393 [1:10:42<5:27:43, 5.43s/it] + Training...: 18% 774/4393 [1:10:47<5:24:56, 5.39s/it] + Training...: 18% 775/4393 [1:10:53<5:23:04, 5.36s/it] + Training...: 18% 776/4393 [1:10:58<5:19:32, 5.30s/it] + Training...: 18% 777/4393 [1:11:03<5:17:54, 5.28s/it] + Training...: 18% 778/4393 [1:11:08<5:15:05, 5.23s/it] + Training...: 18% 779/4393 [1:11:13<5:10:54, 5.16s/it] + Training...: 18% 780/4393 [1:11:18<5:07:22, 5.10s/it] + Training...: 18% 781/4393 [1:11:23<5:04:40, 5.06s/it] + Training...: 18% 782/4393 [1:11:28<5:00:58, 5.00s/it] + Training...: 18% 783/4393 [1:11:33<4:58:47, 4.97s/it] + Training...: 18% 784/4393 [1:11:38<4:54:09, 4.89s/it] + Training...: 18% 785/4393 [1:11:42<4:48:56, 4.80s/it] + Training...: 18% 786/4393 [1:11:47<4:41:15, 4.68s/it] + Training...: 18% 787/4393 [1:11:51<4:35:30, 4.58s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:48:56<17:31:17, 21025.84s/it] + Training...: 18% 787/4393 [1:11:56<4:35:30, 4.58s/it] + Training...: 18% 788/4393 [1:11:56<4:37:36, 4.62s/it] + Training...: 18% 789/4393 [1:12:00<4:25:07, 4.41s/it] + Training...: 18% 790/4393 [1:12:03<4:12:21, 4.20s/it] + Training...: 18% 791/4393 [1:12:07<3:59:09, 3.98s/it] + Training...: 18% 792/4393 [1:12:10<3:45:33, 3.76s/it] + Training...: 18% 793/4393 [1:12:13<3:30:32, 3.51s/it] + Training...: 18% 794/4393 [1:12:16<3:15:21, 3.26s/it] + Training...: 18% 795/4393 [1:12:18<2:58:13, 2.97s/it] + Training...: 18% 796/4393 [1:12:20<2:41:47, 2.70s/it] + Training...: 18% 797/4393 [1:12:22<2:24:26, 2.41s/it] + Training...: 18% 798/4393 [1:12:23<2:07:20, 2.13s/it] + Training...: 18% 799/4393 [1:12:24<1:51:19, 1.86s/it] + Training...: 18% 800/4393 [1:12:25<1:35:38, 1.60s/it] + Training...: 18% 801/4393 [1:12:32<2:59:11, 2.99s/it] + Training...: 18% 802/4393 [1:12:38<3:56:52, 3.96s/it] + Training...: 18% 803/4393 [1:12:44<4:36:43, 4.62s/it] + Training...: 18% 804/4393 [1:12:50<5:01:33, 5.04s/it] + Training...: 18% 805/4393 [1:12:56<5:19:01, 5.33s/it] + Training...: 18% 806/4393 [1:13:02<5:29:19, 5.51s/it] + Training...: 18% 807/4393 [1:13:08<5:37:15, 5.64s/it] + Training...: 18% 808/4393 [1:13:14<5:41:35, 5.72s/it] + Training...: 18% 809/4393 [1:13:20<5:44:11, 5.76s/it] + Training...: 18% 810/4393 [1:13:25<5:44:37, 5.77s/it] + Training...: 18% 811/4393 [1:13:31<5:44:42, 5.77s/it] + Training...: 18% 812/4393 [1:13:37<5:44:28, 5.77s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:50:44<17:31:17, 21025.84s/it] + Training...: 18% 812/4393 [1:13:44<5:44:28, 5.77s/it] + Training...: 19% 813/4393 [1:13:44<5:58:44, 6.01s/it] + Training...: 19% 814/4393 [1:13:49<5:55:32, 5.96s/it] + Training...: 19% 815/4393 [1:13:55<5:51:29, 5.89s/it] + Training...: 19% 816/4393 [1:14:01<5:47:11, 5.82s/it] + Training...: 19% 817/4393 [1:14:06<5:43:12, 5.76s/it] + Training...: 19% 818/4393 [1:14:12<5:38:33, 5.68s/it] + Training...: 19% 819/4393 [1:14:17<5:35:45, 5.64s/it] + Training...: 19% 820/4393 [1:14:23<5:31:39, 5.57s/it] + Training...: 19% 821/4393 [1:14:28<5:29:21, 5.53s/it] + Training...: 19% 822/4393 [1:14:34<5:26:00, 5.48s/it] + Training...: 19% 823/4393 [1:14:39<5:23:48, 5.44s/it] + Training...: 19% 824/4393 [1:14:44<5:21:39, 5.41s/it] + Training...: 19% 825/4393 [1:14:50<5:19:20, 5.37s/it] + Training...: 19% 826/4393 [1:14:55<5:16:22, 5.32s/it] + Training...: 19% 827/4393 [1:15:00<5:14:49, 5.30s/it] + Training...: 19% 828/4393 [1:15:05<5:11:39, 5.25s/it] + Training...: 19% 829/4393 [1:15:10<5:10:14, 5.22s/it] + Training...: 19% 830/4393 [1:15:16<5:08:24, 5.19s/it] + Training...: 19% 831/4393 [1:15:21<5:05:38, 5.15s/it] + Training...: 19% 832/4393 [1:15:25<5:00:31, 5.06s/it] + Training...: 19% 833/4393 [1:15:30<4:57:28, 5.01s/it] + Training...: 19% 834/4393 [1:15:35<4:51:14, 4.91s/it] + Training...: 19% 835/4393 [1:15:40<4:46:59, 4.84s/it] + Training...: 19% 836/4393 [1:15:44<4:43:09, 4.78s/it] + Training...: 19% 837/4393 [1:15:49<4:38:25, 4.70s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:52:54<17:31:17, 21025.84s/it] + Training...: 19% 837/4393 [1:15:54<4:38:25, 4.70s/it] + Training...: 19% 838/4393 [1:15:54<4:40:07, 4.73s/it] + Training...: 19% 839/4393 [1:15:58<4:26:17, 4.50s/it] + Training...: 19% 840/4393 [1:16:01<4:11:56, 4.25s/it] + Training...: 19% 841/4393 [1:16:05<3:57:02, 4.00s/it] + Training...: 19% 842/4393 [1:16:08<3:41:27, 3.74s/it] + Training...: 19% 843/4393 [1:16:11<3:26:03, 3.48s/it] + Training...: 19% 844/4393 [1:16:13<3:11:36, 3.24s/it] + Training...: 19% 845/4393 [1:16:16<2:56:57, 2.99s/it] + Training...: 19% 846/4393 [1:16:18<2:42:00, 2.74s/it] + Training...: 19% 847/4393 [1:16:20<2:26:26, 2.48s/it] + Training...: 19% 848/4393 [1:16:21<2:11:02, 2.22s/it] + Training...: 19% 849/4393 [1:16:23<1:55:08, 1.95s/it] + Training...: 19% 850/4393 [1:16:24<1:39:17, 1.68s/it] + Training...: 19% 851/4393 [1:16:30<3:01:41, 3.08s/it] + Training...: 19% 852/4393 [1:16:36<3:57:48, 4.03s/it] + Training...: 19% 853/4393 [1:16:43<4:35:29, 4.67s/it] + Training...: 19% 854/4393 [1:16:49<4:59:12, 5.07s/it] + Training...: 19% 855/4393 [1:16:55<5:15:28, 5.35s/it] + Training...: 19% 856/4393 [1:17:00<5:25:41, 5.52s/it] + Training...: 20% 857/4393 [1:17:07<5:34:04, 5.67s/it] + Training...: 20% 858/4393 [1:17:12<5:37:43, 5.73s/it] + Training...: 20% 859/4393 [1:17:19<5:45:07, 5.86s/it] + Training...: 20% 860/4393 [1:17:24<5:44:25, 5.85s/it] + Training...: 20% 861/4393 [1:17:30<5:43:29, 5.84s/it] + Training...: 20% 862/4393 [1:17:36<5:41:26, 5.80s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:54:42<17:31:17, 21025.84s/it] + Training...: 20% 862/4393 [1:17:42<5:41:26, 5.80s/it] + Training...: 20% 863/4393 [1:17:42<5:54:24, 6.02s/it] + Training...: 20% 864/4393 [1:17:48<5:48:00, 5.92s/it] + Training...: 20% 865/4393 [1:17:54<5:43:38, 5.84s/it] + Training...: 20% 866/4393 [1:17:59<5:39:37, 5.78s/it] + Training...: 20% 867/4393 [1:18:05<5:37:59, 5.75s/it] + Training...: 20% 868/4393 [1:18:11<5:34:01, 5.69s/it] + Training...: 20% 869/4393 [1:18:16<5:30:49, 5.63s/it] + Training...: 20% 870/4393 [1:18:22<5:27:51, 5.58s/it] + Training...: 20% 871/4393 [1:18:27<5:25:14, 5.54s/it] + Training...: 20% 872/4393 [1:18:32<5:22:55, 5.50s/it] + Training...: 20% 873/4393 [1:18:38<5:22:07, 5.49s/it] + Training...: 20% 874/4393 [1:18:43<5:22:38, 5.50s/it] + Training...: 20% 875/4393 [1:18:49<5:19:26, 5.45s/it] + Training...: 20% 876/4393 [1:18:54<5:14:22, 5.36s/it] + Training...: 20% 877/4393 [1:18:59<5:09:54, 5.29s/it] + Training...: 20% 878/4393 [1:19:04<5:05:59, 5.22s/it] + Training...: 20% 879/4393 [1:19:09<5:03:06, 5.18s/it] + Training...: 20% 880/4393 [1:19:14<4:59:18, 5.11s/it] + Training...: 20% 881/4393 [1:19:19<4:56:23, 5.06s/it] + Training...: 20% 882/4393 [1:19:24<4:52:52, 5.00s/it] + Training...: 20% 883/4393 [1:19:29<4:49:45, 4.95s/it] + Training...: 20% 884/4393 [1:19:33<4:43:30, 4.85s/it] + Training...: 20% 885/4393 [1:19:38<4:38:46, 4.77s/it] + Training...: 20% 886/4393 [1:19:42<4:31:38, 4.65s/it] + Training...: 20% 887/4393 [1:19:47<4:25:05, 4.54s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:56:51<17:31:17, 21025.84s/it] + Training...: 20% 887/4393 [1:19:51<4:25:05, 4.54s/it] + Training...: 20% 888/4393 [1:19:51<4:28:19, 4.59s/it] + Training...: 20% 889/4393 [1:19:55<4:16:22, 4.39s/it] + Training...: 20% 890/4393 [1:19:59<4:03:09, 4.16s/it] + Training...: 20% 891/4393 [1:20:02<3:52:04, 3.98s/it] + Training...: 20% 892/4393 [1:20:06<3:38:59, 3.75s/it] + Training...: 20% 893/4393 [1:20:09<3:24:19, 3.50s/it] + Training...: 20% 894/4393 [1:20:11<3:09:20, 3.25s/it] + Training...: 20% 895/4393 [1:20:14<2:53:47, 2.98s/it] + Training...: 20% 896/4393 [1:20:16<2:38:50, 2.73s/it] + Training...: 20% 897/4393 [1:20:18<2:22:34, 2.45s/it] + Training...: 20% 898/4393 [1:20:19<2:06:16, 2.17s/it] + Training...: 20% 899/4393 [1:20:20<1:50:36, 1.90s/it] + Training...: 20% 900/4393 [1:20:21<1:34:53, 1.63s/it] + Training...: 21% 901/4393 [1:20:28<2:54:33, 3.00s/it] + Training...: 21% 902/4393 [1:20:34<3:50:29, 3.96s/it] + Training...: 21% 903/4393 [1:20:40<4:28:03, 4.61s/it] + Training...: 21% 904/4393 [1:20:46<4:53:17, 5.04s/it] + Training...: 21% 905/4393 [1:20:52<5:09:45, 5.33s/it] + Training...: 21% 906/4393 [1:20:58<5:20:26, 5.51s/it] + Training...: 21% 907/4393 [1:21:04<5:27:06, 5.63s/it] + Training...: 21% 908/4393 [1:21:10<5:31:35, 5.71s/it] + Training...: 21% 909/4393 [1:21:16<5:34:59, 5.77s/it] + Training...: 21% 910/4393 [1:21:21<5:36:25, 5.80s/it] + Training...: 21% 911/4393 [1:21:27<5:41:35, 5.89s/it] + Training...: 21% 912/4393 [1:21:33<5:38:35, 5.84s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [53:58:40<17:31:17, 21025.84s/it] + Training...: 21% 912/4393 [1:21:40<5:38:35, 5.84s/it] + Training...: 21% 913/4393 [1:21:40<5:50:03, 6.04s/it] + Training...: 21% 914/4393 [1:21:45<5:43:46, 5.93s/it] + Training...: 21% 915/4393 [1:21:51<5:40:35, 5.88s/it] + Training...: 21% 916/4393 [1:21:57<5:36:19, 5.80s/it] + Training...: 21% 917/4393 [1:22:03<5:35:41, 5.79s/it] + Training...: 21% 918/4393 [1:22:08<5:31:00, 5.72s/it] + Training...: 21% 919/4393 [1:22:14<5:27:02, 5.65s/it] + Training...: 21% 920/4393 [1:22:19<5:22:50, 5.58s/it] + Training...: 21% 921/4393 [1:22:24<5:19:34, 5.52s/it] + Training...: 21% 922/4393 [1:22:30<5:16:06, 5.46s/it] + Training...: 21% 923/4393 [1:22:35<5:13:40, 5.42s/it] + Training...: 21% 924/4393 [1:22:40<5:11:05, 5.38s/it] + Training...: 21% 925/4393 [1:22:46<5:08:44, 5.34s/it] + Training...: 21% 926/4393 [1:22:51<5:07:25, 5.32s/it] + Training...: 21% 927/4393 [1:22:56<5:05:43, 5.29s/it] + Training...: 21% 928/4393 [1:23:01<5:01:32, 5.22s/it] + Training...: 21% 929/4393 [1:23:06<4:59:14, 5.18s/it] + Training...: 21% 930/4393 [1:23:11<4:54:54, 5.11s/it] + Training...: 21% 931/4393 [1:23:16<4:51:56, 5.06s/it] + Training...: 21% 932/4393 [1:23:21<4:47:38, 4.99s/it] + Training...: 21% 933/4393 [1:23:26<4:44:07, 4.93s/it] + Training...: 21% 934/4393 [1:23:30<4:39:21, 4.85s/it] + Training...: 21% 935/4393 [1:23:35<4:34:18, 4.76s/it] + Training...: 21% 936/4393 [1:23:39<4:27:51, 4.65s/it] + Training...: 21% 937/4393 [1:23:44<4:21:00, 4.53s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:00:48<17:31:17, 21025.84s/it] + Training...: 21% 937/4393 [1:23:48<4:21:00, 4.53s/it] + Training...: 21% 938/4393 [1:23:48<4:23:08, 4.57s/it] + Training...: 21% 939/4393 [1:23:52<4:11:53, 4.38s/it] + Training...: 21% 940/4393 [1:23:56<3:59:22, 4.16s/it] + Training...: 21% 941/4393 [1:23:59<3:47:05, 3.95s/it] + Training...: 21% 942/4393 [1:24:03<3:35:35, 3.75s/it] + Training...: 21% 943/4393 [1:24:05<3:20:54, 3.49s/it] + Training...: 21% 944/4393 [1:24:08<3:07:13, 3.26s/it] + Training...: 22% 945/4393 [1:24:11<2:52:27, 3.00s/it] + Training...: 22% 946/4393 [1:24:13<2:37:57, 2.75s/it] + Training...: 22% 947/4393 [1:24:15<2:23:23, 2.50s/it] + Training...: 22% 948/4393 [1:24:16<2:08:57, 2.25s/it] + Training...: 22% 949/4393 [1:24:18<1:53:50, 1.98s/it] + Training...: 22% 950/4393 [1:24:19<1:37:48, 1.70s/it] + Training...: 22% 951/4393 [1:24:25<2:55:11, 3.05s/it] + Training...: 22% 952/4393 [1:24:31<3:49:54, 4.01s/it] + Training...: 22% 953/4393 [1:24:37<4:24:58, 4.62s/it] + Training...: 22% 954/4393 [1:24:43<4:48:23, 5.03s/it] + Training...: 22% 955/4393 [1:24:49<5:05:52, 5.34s/it] + Training...: 22% 956/4393 [1:24:55<5:14:48, 5.50s/it] + Training...: 22% 957/4393 [1:25:01<5:21:48, 5.62s/it] + Training...: 22% 958/4393 [1:25:07<5:25:46, 5.69s/it] + Training...: 22% 959/4393 [1:25:13<5:27:53, 5.73s/it] + Training...: 22% 960/4393 [1:25:18<5:28:10, 5.74s/it] + Training...: 22% 961/4393 [1:25:24<5:29:18, 5.76s/it] + Training...: 22% 962/4393 [1:25:30<5:28:15, 5.74s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:02:36<17:31:17, 21025.84s/it] + Training...: 22% 962/4393 [1:25:36<5:28:15, 5.74s/it] + Training...: 22% 963/4393 [1:25:36<5:41:28, 5.97s/it] + Training...: 22% 964/4393 [1:25:42<5:36:15, 5.88s/it] + Training...: 22% 965/4393 [1:25:48<5:31:30, 5.80s/it] + Training...: 22% 966/4393 [1:25:53<5:27:44, 5.74s/it] + Training...: 22% 967/4393 [1:25:59<5:24:46, 5.69s/it] + Training...: 22% 968/4393 [1:26:04<5:20:40, 5.62s/it] + Training...: 22% 969/4393 [1:26:10<5:18:06, 5.57s/it] + Training...: 22% 970/4393 [1:26:15<5:15:08, 5.52s/it] + Training...: 22% 971/4393 [1:26:21<5:14:33, 5.52s/it] + Training...: 22% 972/4393 [1:26:26<5:13:36, 5.50s/it] + Training...: 22% 973/4393 [1:26:32<5:12:07, 5.48s/it] + Training...: 22% 974/4393 [1:26:37<5:08:56, 5.42s/it] + Training...: 22% 975/4393 [1:26:42<5:05:11, 5.36s/it] + Training...: 22% 976/4393 [1:26:47<5:01:07, 5.29s/it] + Training...: 22% 977/4393 [1:26:52<4:58:43, 5.25s/it] + Training...: 22% 978/4393 [1:26:57<4:54:49, 5.18s/it] + Training...: 22% 979/4393 [1:27:02<4:52:21, 5.14s/it] + Training...: 22% 980/4393 [1:27:07<4:48:21, 5.07s/it] + Training...: 22% 981/4393 [1:27:12<4:45:33, 5.02s/it] + Training...: 22% 982/4393 [1:27:17<4:41:11, 4.95s/it] + Training...: 22% 983/4393 [1:27:22<4:37:17, 4.88s/it] + Training...: 22% 984/4393 [1:27:26<4:33:00, 4.80s/it] + Training...: 22% 985/4393 [1:27:31<4:29:17, 4.74s/it] + Training...: 22% 986/4393 [1:27:35<4:23:27, 4.64s/it] + Training...: 22% 987/4393 [1:27:40<4:16:34, 4.52s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:04:44<17:31:17, 21025.84s/it] + Training...: 22% 987/4393 [1:27:44<4:16:34, 4.52s/it] + Training...: 22% 988/4393 [1:27:44<4:19:00, 4.56s/it] + Training...: 23% 989/4393 [1:27:48<4:07:33, 4.36s/it] + Training...: 23% 990/4393 [1:27:52<3:55:03, 4.14s/it] + Training...: 23% 991/4393 [1:27:55<3:42:57, 3.93s/it] + Training...: 23% 992/4393 [1:27:58<3:30:41, 3.72s/it] + Training...: 23% 993/4393 [1:28:01<3:16:29, 3.47s/it] + Training...: 23% 994/4393 [1:28:04<3:03:01, 3.23s/it] + Training...: 23% 995/4393 [1:28:06<2:48:17, 2.97s/it] + Training...: 23% 996/4393 [1:28:08<2:32:43, 2.70s/it] + Training...: 23% 997/4393 [1:28:10<2:18:02, 2.44s/it] + Training...: 23% 998/4393 [1:28:12<2:04:51, 2.21s/it] + Training...: 23% 999/4393 [1:28:13<1:49:06, 1.93s/it] + Training...: 23% 1000/4393 [1:28:14<1:33:18, 1.65s/it] + Training...: 23% 1001/4393 [1:28:21<2:53:09, 3.06s/it] + Training...: 23% 1002/4393 [1:28:27<3:48:01, 4.03s/it] + Training...: 23% 1003/4393 [1:28:33<4:22:15, 4.64s/it] + Training...: 23% 1004/4393 [1:28:39<4:44:40, 5.04s/it] + Training...: 23% 1005/4393 [1:28:45<4:59:57, 5.31s/it] + Training...: 23% 1006/4393 [1:28:51<5:10:28, 5.50s/it] + Training...: 23% 1007/4393 [1:28:57<5:17:39, 5.63s/it] + Training...: 23% 1008/4393 [1:29:03<5:20:37, 5.68s/it] + Training...: 23% 1009/4393 [1:29:08<5:22:44, 5.72s/it] + Training...: 23% 1010/4393 [1:29:14<5:24:21, 5.75s/it] + Training...: 23% 1011/4393 [1:29:20<5:24:38, 5.76s/it] + Training...: 23% 1012/4393 [1:29:26<5:24:43, 5.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:06:32<17:31:17, 21025.84s/it] + Training...: 23% 1012/4393 [1:29:32<5:24:43, 5.76s/it] + Training...: 23% 1013/4393 [1:29:32<5:38:07, 6.00s/it] + Training...: 23% 1014/4393 [1:29:38<5:32:35, 5.91s/it] + Training...: 23% 1015/4393 [1:29:44<5:28:00, 5.83s/it] + Training...: 23% 1016/4393 [1:29:49<5:25:57, 5.79s/it] + Training...: 23% 1017/4393 [1:29:55<5:26:38, 5.81s/it] + Training...: 23% 1018/4393 [1:30:01<5:23:14, 5.75s/it] + Training...: 23% 1019/4393 [1:30:06<5:20:14, 5.69s/it] + Training...: 23% 1020/4393 [1:30:12<5:17:32, 5.65s/it] + Training...: 23% 1021/4393 [1:30:17<5:14:25, 5.59s/it] + Training...: 23% 1022/4393 [1:30:23<5:10:34, 5.53s/it] + Training...: 23% 1023/4393 [1:30:28<5:08:27, 5.49s/it] + Training...: 23% 1024/4393 [1:30:34<5:06:28, 5.46s/it] + Training...: 23% 1025/4393 [1:30:39<5:04:20, 5.42s/it] + Training...: 23% 1026/4393 [1:30:44<5:01:45, 5.38s/it] + Training...: 23% 1027/4393 [1:30:49<5:00:10, 5.35s/it] + Training...: 23% 1028/4393 [1:30:55<4:56:17, 5.28s/it] + Training...: 23% 1029/4393 [1:31:00<4:54:32, 5.25s/it] + Training...: 23% 1030/4393 [1:31:05<4:51:09, 5.19s/it] + Training...: 23% 1031/4393 [1:31:10<4:50:44, 5.19s/it] + Training...: 23% 1032/4393 [1:31:15<4:47:53, 5.14s/it] + Training...: 24% 1033/4393 [1:31:20<4:49:20, 5.17s/it] + Training...: 24% 1034/4393 [1:31:25<4:45:54, 5.11s/it] + Training...: 24% 1035/4393 [1:31:30<4:41:04, 5.02s/it] + Training...: 24% 1036/4393 [1:31:35<4:35:02, 4.92s/it] + Training...: 24% 1037/4393 [1:31:39<4:27:45, 4.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:08:44<17:31:17, 21025.84s/it] + Training...: 24% 1037/4393 [1:31:44<4:27:45, 4.79s/it] + Training...: 24% 1038/4393 [1:31:44<4:31:16, 4.85s/it] + Training...: 24% 1039/4393 [1:31:48<4:15:36, 4.57s/it] + Training...: 24% 1040/4393 [1:31:52<3:59:44, 4.29s/it] + Training...: 24% 1041/4393 [1:31:55<3:45:38, 4.04s/it] + Training...: 24% 1042/4393 [1:31:58<3:31:04, 3.78s/it] + Training...: 24% 1043/4393 [1:32:01<3:16:49, 3.53s/it] + Training...: 24% 1044/4393 [1:32:04<3:03:30, 3.29s/it] + Training...: 24% 1045/4393 [1:32:06<2:48:56, 3.03s/it] + Training...: 24% 1046/4393 [1:32:09<2:34:09, 2.76s/it] + Training...: 24% 1047/4393 [1:32:10<2:19:33, 2.50s/it] + Training...: 24% 1048/4393 [1:32:12<2:04:30, 2.23s/it] + Training...: 24% 1049/4393 [1:32:13<1:49:50, 1.97s/it] + Training...: 24% 1050/4393 [1:32:14<1:34:15, 1.69s/it] + Training...: 24% 1051/4393 [1:32:21<2:53:55, 3.12s/it] + Training...: 24% 1052/4393 [1:32:27<3:46:24, 4.07s/it] + Training...: 24% 1053/4393 [1:32:33<4:19:55, 4.67s/it] + Training...: 24% 1054/4393 [1:32:39<4:41:33, 5.06s/it] + Training...: 24% 1055/4393 [1:32:45<4:57:48, 5.35s/it] + Training...: 24% 1056/4393 [1:32:51<5:08:00, 5.54s/it] + Training...: 24% 1057/4393 [1:32:57<5:15:38, 5.68s/it] + Training...: 24% 1058/4393 [1:33:03<5:19:06, 5.74s/it] + Training...: 24% 1059/4393 [1:33:09<5:21:05, 5.78s/it] + Training...: 24% 1060/4393 [1:33:15<5:21:30, 5.79s/it] + Training...: 24% 1061/4393 [1:33:21<5:23:32, 5.83s/it] + Training...: 24% 1062/4393 [1:33:27<5:25:21, 5.86s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:10:33<17:31:17, 21025.84s/it] + Training...: 24% 1062/4393 [1:33:33<5:25:21, 5.86s/it] + Training...: 24% 1063/4393 [1:33:33<5:38:01, 6.09s/it] + Training...: 24% 1064/4393 [1:33:39<5:30:30, 5.96s/it] + Training...: 24% 1065/4393 [1:33:45<5:26:05, 5.88s/it] + Training...: 24% 1066/4393 [1:33:50<5:22:14, 5.81s/it] + Training...: 24% 1067/4393 [1:33:56<5:22:41, 5.82s/it] + Training...: 24% 1068/4393 [1:34:02<5:18:14, 5.74s/it] + Training...: 24% 1069/4393 [1:34:07<5:13:57, 5.67s/it] + Training...: 24% 1070/4393 [1:34:13<5:11:15, 5.62s/it] + Training...: 24% 1071/4393 [1:34:18<5:07:39, 5.56s/it] + Training...: 24% 1072/4393 [1:34:23<5:04:18, 5.50s/it] + Training...: 24% 1073/4393 [1:34:29<5:01:20, 5.45s/it] + Training...: 24% 1074/4393 [1:34:34<5:00:15, 5.43s/it] + Training...: 24% 1075/4393 [1:34:39<4:57:30, 5.38s/it] + Training...: 24% 1076/4393 [1:34:45<4:54:38, 5.33s/it] + Training...: 25% 1077/4393 [1:34:50<4:51:53, 5.28s/it] + Training...: 25% 1078/4393 [1:34:55<4:48:54, 5.23s/it] + Training...: 25% 1079/4393 [1:35:00<4:46:42, 5.19s/it] + Training...: 25% 1080/4393 [1:35:05<4:43:23, 5.13s/it] + Training...: 25% 1081/4393 [1:35:10<4:41:01, 5.09s/it] + Training...: 25% 1082/4393 [1:35:15<4:38:19, 5.04s/it] + Training...: 25% 1083/4393 [1:35:20<4:35:22, 4.99s/it] + Training...: 25% 1084/4393 [1:35:25<4:31:24, 4.92s/it] + Training...: 25% 1085/4393 [1:35:29<4:27:45, 4.86s/it] + Training...: 25% 1086/4393 [1:35:34<4:24:03, 4.79s/it] + Training...: 25% 1087/4393 [1:35:38<4:18:25, 4.69s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:12:43<17:31:17, 21025.84s/it] + Training...: 25% 1087/4393 [1:35:43<4:18:25, 4.69s/it] + Training...: 25% 1088/4393 [1:35:43<4:20:25, 4.73s/it] + Training...: 25% 1089/4393 [1:35:47<4:08:12, 4.51s/it] + Training...: 25% 1090/4393 [1:35:51<3:54:52, 4.27s/it] + Training...: 25% 1091/4393 [1:35:54<3:41:36, 4.03s/it] + Training...: 25% 1092/4393 [1:35:58<3:28:07, 3.78s/it] + Training...: 25% 1093/4393 [1:36:01<3:13:49, 3.52s/it] + Training...: 25% 1094/4393 [1:36:03<3:00:19, 3.28s/it] + Training...: 25% 1095/4393 [1:36:06<2:45:13, 3.01s/it] + Training...: 25% 1096/4393 [1:36:08<2:30:25, 2.74s/it] + Training...: 25% 1097/4393 [1:36:10<2:15:29, 2.47s/it] + Training...: 25% 1098/4393 [1:36:11<2:00:11, 2.19s/it] + Training...: 25% 1099/4393 [1:36:12<1:45:21, 1.92s/it] + Training...: 25% 1100/4393 [1:36:13<1:30:24, 1.65s/it] + Training...: 25% 1101/4393 [1:36:20<2:44:54, 3.01s/it] + Training...: 25% 1102/4393 [1:36:26<3:38:18, 3.98s/it] + Training...: 25% 1103/4393 [1:36:32<4:13:12, 4.62s/it] + Training...: 25% 1104/4393 [1:36:38<4:35:44, 5.03s/it] + Training...: 25% 1105/4393 [1:36:44<4:50:45, 5.31s/it] + Training...: 25% 1106/4393 [1:36:50<5:01:37, 5.51s/it] + Training...: 25% 1107/4393 [1:36:56<5:10:23, 5.67s/it] + Training...: 25% 1108/4393 [1:37:02<5:16:47, 5.79s/it] + Training...: 25% 1109/4393 [1:37:08<5:20:02, 5.85s/it] + Training...: 25% 1110/4393 [1:37:14<5:18:04, 5.81s/it] + Training...: 25% 1111/4393 [1:37:19<5:16:50, 5.79s/it] + Training...: 25% 1112/4393 [1:37:25<5:14:44, 5.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:14:32<17:31:17, 21025.84s/it] + Training...: 25% 1112/4393 [1:37:32<5:14:44, 5.76s/it] + Training...: 25% 1113/4393 [1:37:32<5:27:23, 5.99s/it] + Training...: 25% 1114/4393 [1:37:37<5:21:57, 5.89s/it] + Training...: 25% 1115/4393 [1:37:43<5:17:51, 5.82s/it] + Training...: 25% 1116/4393 [1:37:49<5:14:48, 5.76s/it] + Training...: 25% 1117/4393 [1:37:54<5:12:07, 5.72s/it] + Training...: 25% 1118/4393 [1:38:00<5:08:31, 5.65s/it] + Training...: 25% 1119/4393 [1:38:05<5:06:55, 5.62s/it] + Training...: 25% 1120/4393 [1:38:11<5:05:04, 5.59s/it] + Training...: 26% 1121/4393 [1:38:16<5:03:28, 5.56s/it] + Training...: 26% 1122/4393 [1:38:22<5:01:21, 5.53s/it] + Training...: 26% 1123/4393 [1:38:27<4:59:53, 5.50s/it] + Training...: 26% 1124/4393 [1:38:33<4:58:57, 5.49s/it] + Training...: 26% 1125/4393 [1:38:38<4:58:55, 5.49s/it] + Training...: 26% 1126/4393 [1:38:43<4:56:33, 5.45s/it] + Training...: 26% 1127/4393 [1:38:49<4:54:08, 5.40s/it] + Training...: 26% 1128/4393 [1:38:54<4:52:50, 5.38s/it] + Training...: 26% 1129/4393 [1:38:59<4:51:21, 5.36s/it] + Training...: 26% 1130/4393 [1:39:05<4:47:49, 5.29s/it] + Training...: 26% 1131/4393 [1:39:10<4:45:17, 5.25s/it] + Training...: 26% 1132/4393 [1:39:15<4:43:14, 5.21s/it] + Training...: 26% 1133/4393 [1:39:20<4:40:06, 5.16s/it] + Training...: 26% 1134/4393 [1:39:25<4:34:53, 5.06s/it] + Training...: 26% 1135/4393 [1:39:29<4:29:55, 4.97s/it] + Training...: 26% 1136/4393 [1:39:34<4:24:02, 4.86s/it] + Training...: 26% 1137/4393 [1:39:39<4:18:04, 4.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:16:43<17:31:17, 21025.84s/it] + Training...: 26% 1137/4393 [1:39:43<4:18:04, 4.76s/it] + Training...: 26% 1138/4393 [1:39:43<4:20:21, 4.80s/it] + Training...: 26% 1139/4393 [1:39:47<4:06:34, 4.55s/it] + Training...: 26% 1140/4393 [1:39:51<3:52:32, 4.29s/it] + Training...: 26% 1141/4393 [1:39:55<3:38:43, 4.04s/it] + Training...: 26% 1142/4393 [1:39:58<3:26:12, 3.81s/it] + Training...: 26% 1143/4393 [1:40:01<3:12:37, 3.56s/it] + Training...: 26% 1144/4393 [1:40:03<2:58:45, 3.30s/it] + Training...: 26% 1145/4393 [1:40:06<2:43:36, 3.02s/it] + Training...: 26% 1146/4393 [1:40:08<2:28:54, 2.75s/it] + Training...: 26% 1147/4393 [1:40:10<2:13:36, 2.47s/it] + Training...: 26% 1148/4393 [1:40:11<1:58:22, 2.19s/it] + Training...: 26% 1149/4393 [1:40:13<1:44:13, 1.93s/it] + Training...: 26% 1150/4393 [1:40:14<1:29:39, 1.66s/it] + Training...: 26% 1151/4393 [1:40:20<2:43:27, 3.03s/it] + Training...: 26% 1152/4393 [1:40:26<3:35:26, 3.99s/it] + Training...: 26% 1153/4393 [1:40:32<4:10:46, 4.64s/it] + Training...: 26% 1154/4393 [1:40:39<4:37:32, 5.14s/it] + Training...: 26% 1155/4393 [1:40:45<4:50:26, 5.38s/it] + Training...: 26% 1156/4393 [1:40:50<4:59:23, 5.55s/it] + Training...: 26% 1157/4393 [1:40:56<5:05:40, 5.67s/it] + Training...: 26% 1158/4393 [1:41:02<5:08:23, 5.72s/it] + Training...: 26% 1159/4393 [1:41:08<5:10:28, 5.76s/it] + Training...: 26% 1160/4393 [1:41:14<5:10:57, 5.77s/it] + Training...: 26% 1161/4393 [1:41:20<5:10:55, 5.77s/it] + Training...: 26% 1162/4393 [1:41:25<5:09:40, 5.75s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:18:32<17:31:17, 21025.84s/it] + Training...: 26% 1162/4393 [1:41:32<5:09:40, 5.75s/it] + Training...: 26% 1163/4393 [1:41:32<5:23:21, 6.01s/it] + Training...: 26% 1164/4393 [1:41:38<5:17:48, 5.91s/it] + Training...: 27% 1165/4393 [1:41:43<5:14:50, 5.85s/it] + Training...: 27% 1166/4393 [1:41:49<5:10:50, 5.78s/it] + Training...: 27% 1167/4393 [1:41:55<5:07:19, 5.72s/it] + Training...: 27% 1168/4393 [1:42:00<5:05:24, 5.68s/it] + Training...: 27% 1169/4393 [1:42:06<5:07:22, 5.72s/it] + Training...: 27% 1170/4393 [1:42:12<5:05:30, 5.69s/it] + Training...: 27% 1171/4393 [1:42:17<5:06:02, 5.70s/it] + Training...: 27% 1172/4393 [1:42:23<5:02:57, 5.64s/it] + Training...: 27% 1173/4393 [1:42:28<5:01:00, 5.61s/it] + Training...: 27% 1174/4393 [1:42:34<4:58:02, 5.56s/it] + Training...: 27% 1175/4393 [1:42:39<4:55:45, 5.51s/it] + Training...: 27% 1176/4393 [1:42:45<4:53:25, 5.47s/it] + Training...: 27% 1177/4393 [1:42:50<4:51:54, 5.45s/it] + Training...: 27% 1178/4393 [1:42:55<4:49:14, 5.40s/it] + Training...: 27% 1179/4393 [1:43:00<4:46:35, 5.35s/it] + Training...: 27% 1180/4393 [1:43:06<4:42:51, 5.28s/it] + Training...: 27% 1181/4393 [1:43:11<4:39:48, 5.23s/it] + Training...: 27% 1182/4393 [1:43:16<4:36:43, 5.17s/it] + Training...: 27% 1183/4393 [1:43:21<4:34:07, 5.12s/it] + Training...: 27% 1184/4393 [1:43:26<4:29:02, 5.03s/it] + Training...: 27% 1185/4393 [1:43:30<4:25:36, 4.97s/it] + Training...: 27% 1186/4393 [1:43:35<4:23:09, 4.92s/it] + Training...: 27% 1187/4393 [1:43:40<4:16:01, 4.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:20:44<17:31:17, 21025.84s/it] + Training...: 27% 1187/4393 [1:43:45<4:16:01, 4.79s/it] + Training...: 27% 1188/4393 [1:43:45<4:16:53, 4.81s/it] + Training...: 27% 1189/4393 [1:43:48<4:02:16, 4.54s/it] + Training...: 27% 1190/4393 [1:43:52<3:48:20, 4.28s/it] + Training...: 27% 1191/4393 [1:43:56<3:35:35, 4.04s/it] + Training...: 27% 1192/4393 [1:43:59<3:23:12, 3.81s/it] + Training...: 27% 1193/4393 [1:44:02<3:09:04, 3.55s/it] + Training...: 27% 1194/4393 [1:44:04<2:53:37, 3.26s/it] + Training...: 27% 1195/4393 [1:44:07<2:38:07, 2.97s/it] + Training...: 27% 1196/4393 [1:44:09<2:23:20, 2.69s/it] + Training...: 27% 1197/4393 [1:44:11<2:09:28, 2.43s/it] + Training...: 27% 1198/4393 [1:44:12<1:55:31, 2.17s/it] + Training...: 27% 1199/4393 [1:44:13<1:41:40, 1.91s/it] + Training...: 27% 1200/4393 [1:44:14<1:27:53, 1.65s/it] + Training...: 27% 1201/4393 [1:44:21<2:40:34, 3.02s/it] + Training...: 27% 1202/4393 [1:44:27<3:32:11, 3.99s/it] + Training...: 27% 1203/4393 [1:44:33<4:06:02, 4.63s/it] + Training...: 27% 1204/4393 [1:44:39<4:28:27, 5.05s/it] + Training...: 27% 1205/4393 [1:44:45<4:43:24, 5.33s/it] + Training...: 27% 1206/4393 [1:44:51<4:53:14, 5.52s/it] + Training...: 27% 1207/4393 [1:44:57<5:00:55, 5.67s/it] + Training...: 27% 1208/4393 [1:45:03<5:05:02, 5.75s/it] + Training...: 28% 1209/4393 [1:45:09<5:07:18, 5.79s/it] + Training...: 28% 1210/4393 [1:45:15<5:11:36, 5.87s/it] + Training...: 28% 1211/4393 [1:45:21<5:09:13, 5.83s/it] + Training...: 28% 1212/4393 [1:45:26<5:06:40, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:22:33<17:31:17, 21025.84s/it] + Training...: 28% 1212/4393 [1:45:33<5:06:40, 5.78s/it] + Training...: 28% 1213/4393 [1:45:33<5:18:13, 6.00s/it] + Training...: 28% 1214/4393 [1:45:38<5:11:57, 5.89s/it] + Training...: 28% 1215/4393 [1:45:44<5:08:35, 5.83s/it] + Training...: 28% 1216/4393 [1:45:50<5:04:54, 5.76s/it] + Training...: 28% 1217/4393 [1:45:55<5:01:29, 5.70s/it] + Training...: 28% 1218/4393 [1:46:01<4:57:44, 5.63s/it] + Training...: 28% 1219/4393 [1:46:06<4:56:47, 5.61s/it] + Training...: 28% 1220/4393 [1:46:12<4:55:11, 5.58s/it] + Training...: 28% 1221/4393 [1:46:17<4:53:31, 5.55s/it] + Training...: 28% 1222/4393 [1:46:23<4:50:31, 5.50s/it] + Training...: 28% 1223/4393 [1:46:28<4:47:13, 5.44s/it] + Training...: 28% 1224/4393 [1:46:33<4:44:56, 5.39s/it] + Training...: 28% 1225/4393 [1:46:39<4:42:41, 5.35s/it] + Training...: 28% 1226/4393 [1:46:44<4:40:24, 5.31s/it] + Training...: 28% 1227/4393 [1:46:49<4:39:38, 5.30s/it] + Training...: 28% 1228/4393 [1:46:54<4:39:02, 5.29s/it] + Training...: 28% 1229/4393 [1:46:59<4:35:05, 5.22s/it] + Training...: 28% 1230/4393 [1:47:04<4:30:53, 5.14s/it] + Training...: 28% 1231/4393 [1:47:09<4:26:58, 5.07s/it] + Training...: 28% 1232/4393 [1:47:14<4:22:25, 4.98s/it] + Training...: 28% 1233/4393 [1:47:19<4:18:19, 4.90s/it] + Training...: 28% 1234/4393 [1:47:23<4:13:46, 4.82s/it] + Training...: 28% 1235/4393 [1:47:28<4:08:55, 4.73s/it] + Training...: 28% 1236/4393 [1:47:32<4:02:53, 4.62s/it] + Training...: 28% 1237/4393 [1:47:36<3:55:52, 4.48s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:24:41<17:31:17, 21025.84s/it] + Training...: 28% 1237/4393 [1:47:41<3:55:52, 4.48s/it] + Training...: 28% 1238/4393 [1:47:41<3:58:50, 4.54s/it] + Training...: 28% 1239/4393 [1:47:45<3:48:07, 4.34s/it] + Training...: 28% 1240/4393 [1:47:48<3:35:43, 4.10s/it] + Training...: 28% 1241/4393 [1:47:52<3:22:50, 3.86s/it] + Training...: 28% 1242/4393 [1:47:55<3:09:32, 3.61s/it] + Training...: 28% 1243/4393 [1:47:58<2:56:02, 3.35s/it] + Training...: 28% 1244/4393 [1:48:00<2:41:57, 3.09s/it] + Training...: 28% 1245/4393 [1:48:02<2:29:14, 2.84s/it] + Training...: 28% 1246/4393 [1:48:04<2:15:43, 2.59s/it] + Training...: 28% 1247/4393 [1:48:06<2:02:35, 2.34s/it] + Training...: 28% 1248/4393 [1:48:08<1:49:27, 2.09s/it] + Training...: 28% 1249/4393 [1:48:09<1:36:14, 1.84s/it] + Training...: 28% 1250/4393 [1:48:10<1:22:58, 1.58s/it] + Training...: 28% 1251/4393 [1:48:16<2:35:31, 2.97s/it] + Training...: 28% 1252/4393 [1:48:22<3:25:27, 3.92s/it] + Training...: 29% 1253/4393 [1:48:28<3:58:29, 4.56s/it] + Training...: 29% 1254/4393 [1:48:34<4:21:21, 5.00s/it] + Training...: 29% 1255/4393 [1:48:40<4:36:00, 5.28s/it] + Training...: 29% 1256/4393 [1:48:46<4:46:09, 5.47s/it] + Training...: 29% 1257/4393 [1:48:52<4:53:29, 5.62s/it] + Training...: 29% 1258/4393 [1:48:58<4:58:39, 5.72s/it] + Training...: 29% 1259/4393 [1:49:04<5:01:43, 5.78s/it] + Training...: 29% 1260/4393 [1:49:10<5:01:19, 5.77s/it] + Training...: 29% 1261/4393 [1:49:16<5:03:25, 5.81s/it] + Training...: 29% 1262/4393 [1:49:21<5:03:35, 5.82s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:26:28<17:31:17, 21025.84s/it] + Training...: 29% 1262/4393 [1:49:28<5:03:35, 5.82s/it] + Training...: 29% 1263/4393 [1:49:28<5:14:24, 6.03s/it] + Training...: 29% 1264/4393 [1:49:34<5:07:59, 5.91s/it] + Training...: 29% 1265/4393 [1:49:39<5:03:25, 5.82s/it] + Training...: 29% 1266/4393 [1:49:45<4:59:05, 5.74s/it] + Training...: 29% 1267/4393 [1:49:50<4:57:25, 5.71s/it] + Training...: 29% 1268/4393 [1:49:56<4:53:41, 5.64s/it] + Training...: 29% 1269/4393 [1:50:01<4:51:35, 5.60s/it] + Training...: 29% 1270/4393 [1:50:07<4:49:20, 5.56s/it] + Training...: 29% 1271/4393 [1:50:12<4:47:24, 5.52s/it] + Training...: 29% 1272/4393 [1:50:18<4:44:55, 5.48s/it] + Training...: 29% 1273/4393 [1:50:23<4:43:05, 5.44s/it] + Training...: 29% 1274/4393 [1:50:28<4:39:56, 5.39s/it] + Training...: 29% 1275/4393 [1:50:34<4:38:43, 5.36s/it] + Training...: 29% 1276/4393 [1:50:39<4:36:11, 5.32s/it] + Training...: 29% 1277/4393 [1:50:44<4:33:32, 5.27s/it] + Training...: 29% 1278/4393 [1:50:49<4:34:22, 5.28s/it] + Training...: 29% 1279/4393 [1:50:54<4:31:07, 5.22s/it] + Training...: 29% 1280/4393 [1:50:59<4:26:38, 5.14s/it] + Training...: 29% 1281/4393 [1:51:04<4:23:34, 5.08s/it] + Training...: 29% 1282/4393 [1:51:09<4:19:31, 5.01s/it] + Training...: 29% 1283/4393 [1:51:14<4:15:49, 4.94s/it] + Training...: 29% 1284/4393 [1:51:18<4:11:50, 4.86s/it] + Training...: 29% 1285/4393 [1:51:23<4:07:52, 4.79s/it] + Training...: 29% 1286/4393 [1:51:27<4:01:42, 4.67s/it] + Training...: 29% 1287/4393 [1:51:32<3:56:37, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:28:36<17:31:17, 21025.84s/it] + Training...: 29% 1287/4393 [1:51:37<3:56:37, 4.57s/it] + Training...: 29% 1288/4393 [1:51:37<3:59:11, 4.62s/it] + Training...: 29% 1289/4393 [1:51:41<3:49:00, 4.43s/it] + Training...: 29% 1290/4393 [1:51:44<3:36:47, 4.19s/it] + Training...: 29% 1291/4393 [1:51:48<3:24:59, 3.97s/it] + Training...: 29% 1292/4393 [1:51:51<3:13:00, 3.73s/it] + Training...: 29% 1293/4393 [1:51:54<2:59:31, 3.47s/it] + Training...: 29% 1294/4393 [1:51:56<2:45:39, 3.21s/it] + Training...: 29% 1295/4393 [1:51:58<2:30:57, 2.92s/it] + Training...: 30% 1296/4393 [1:52:01<2:17:52, 2.67s/it] + Training...: 30% 1297/4393 [1:52:02<2:04:46, 2.42s/it] + Training...: 30% 1298/4393 [1:52:04<1:51:07, 2.15s/it] + Training...: 30% 1299/4393 [1:52:05<1:37:13, 1.89s/it] + Training...: 30% 1300/4393 [1:52:06<1:23:18, 1.62s/it] + Training...: 30% 1301/4393 [1:52:13<2:36:18, 3.03s/it] + Training...: 30% 1302/4393 [1:52:19<3:26:25, 4.01s/it] + Training...: 30% 1303/4393 [1:52:25<3:58:24, 4.63s/it] + Training...: 30% 1304/4393 [1:52:31<4:19:09, 5.03s/it] + Training...: 30% 1305/4393 [1:52:37<4:33:12, 5.31s/it] + Training...: 30% 1306/4393 [1:52:43<4:41:10, 5.47s/it] + Training...: 30% 1307/4393 [1:52:49<4:47:38, 5.59s/it] + Training...: 30% 1308/4393 [1:52:54<4:51:10, 5.66s/it] + Training...: 30% 1309/4393 [1:53:00<4:54:06, 5.72s/it] + Training...: 30% 1310/4393 [1:53:06<4:54:27, 5.73s/it] + Training...: 30% 1311/4393 [1:53:12<4:54:23, 5.73s/it] + Training...: 30% 1312/4393 [1:53:17<4:52:59, 5.71s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:30:24<17:31:17, 21025.84s/it] + Training...: 30% 1312/4393 [1:53:24<4:52:59, 5.71s/it] + Training...: 30% 1313/4393 [1:53:24<5:06:11, 5.96s/it] + Training...: 30% 1314/4393 [1:53:30<5:00:51, 5.86s/it] + Training...: 30% 1315/4393 [1:53:35<4:57:22, 5.80s/it] + Training...: 30% 1316/4393 [1:53:41<4:53:18, 5.72s/it] + Training...: 30% 1317/4393 [1:53:46<4:51:42, 5.69s/it] + Training...: 30% 1318/4393 [1:53:52<4:49:04, 5.64s/it] + Training...: 30% 1319/4393 [1:53:57<4:47:34, 5.61s/it] + Training...: 30% 1320/4393 [1:54:03<4:44:32, 5.56s/it] + Training...: 30% 1321/4393 [1:54:08<4:43:09, 5.53s/it] + Training...: 30% 1322/4393 [1:54:14<4:40:48, 5.49s/it] + Training...: 30% 1323/4393 [1:54:19<4:38:32, 5.44s/it] + Training...: 30% 1324/4393 [1:54:24<4:36:27, 5.40s/it] + Training...: 30% 1325/4393 [1:54:30<4:34:21, 5.37s/it] + Training...: 30% 1326/4393 [1:54:35<4:32:14, 5.33s/it] + Training...: 30% 1327/4393 [1:54:40<4:30:25, 5.29s/it] + Training...: 30% 1328/4393 [1:54:45<4:27:55, 5.24s/it] + Training...: 30% 1329/4393 [1:54:50<4:24:30, 5.18s/it] + Training...: 30% 1330/4393 [1:54:55<4:20:24, 5.10s/it] + Training...: 30% 1331/4393 [1:55:00<4:16:27, 5.03s/it] + Training...: 30% 1332/4393 [1:55:05<4:11:54, 4.94s/it] + Training...: 30% 1333/4393 [1:55:09<4:07:52, 4.86s/it] + Training...: 30% 1334/4393 [1:55:14<4:02:43, 4.76s/it] + Training...: 30% 1335/4393 [1:55:18<3:57:36, 4.66s/it] + Training...: 30% 1336/4393 [1:55:23<3:51:15, 4.54s/it] + Training...: 30% 1337/4393 [1:55:27<3:45:07, 4.42s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:32:31<17:31:17, 21025.84s/it] + Training...: 30% 1337/4393 [1:55:31<3:45:07, 4.42s/it] + Training...: 30% 1338/4393 [1:55:31<3:46:41, 4.45s/it] + Training...: 30% 1339/4393 [1:55:35<3:35:52, 4.24s/it] + Training...: 31% 1340/4393 [1:55:39<3:24:03, 4.01s/it] + Training...: 31% 1341/4393 [1:55:42<3:12:54, 3.79s/it] + Training...: 31% 1342/4393 [1:55:45<3:02:18, 3.59s/it] + Training...: 31% 1343/4393 [1:55:48<2:51:08, 3.37s/it] + Training...: 31% 1344/4393 [1:55:50<2:36:49, 3.09s/it] + Training...: 31% 1345/4393 [1:55:52<2:23:03, 2.82s/it] + Training...: 31% 1346/4393 [1:55:54<2:09:00, 2.54s/it] + Training...: 31% 1347/4393 [1:55:56<1:55:57, 2.28s/it] + Training...: 31% 1348/4393 [1:55:57<1:43:34, 2.04s/it] + Training...: 31% 1349/4393 [1:55:59<1:31:03, 1.79s/it] + Training...: 31% 1350/4393 [1:56:00<1:18:56, 1.56s/it] + Training...: 31% 1351/4393 [1:56:06<2:30:34, 2.97s/it] + Training...: 31% 1352/4393 [1:56:12<3:19:09, 3.93s/it] + Training...: 31% 1353/4393 [1:56:18<3:52:54, 4.60s/it] + Training...: 31% 1354/4393 [1:56:24<4:13:41, 5.01s/it] + Training...: 31% 1355/4393 [1:56:30<4:29:31, 5.32s/it] + Training...: 31% 1356/4393 [1:56:36<4:42:42, 5.59s/it] + Training...: 31% 1357/4393 [1:56:42<4:47:37, 5.68s/it] + Training...: 31% 1358/4393 [1:56:48<4:49:36, 5.73s/it] + Training...: 31% 1359/4393 [1:56:54<4:51:23, 5.76s/it] + Training...: 31% 1360/4393 [1:57:00<4:51:29, 5.77s/it] + Training...: 31% 1361/4393 [1:57:06<4:51:04, 5.76s/it] + Training...: 31% 1362/4393 [1:57:11<4:49:57, 5.74s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:34:18<17:31:17, 21025.84s/it] + Training...: 31% 1362/4393 [1:57:18<4:49:57, 5.74s/it] + Training...: 31% 1363/4393 [1:57:18<5:01:26, 5.97s/it] + Training...: 31% 1364/4393 [1:57:23<4:56:26, 5.87s/it] + Training...: 31% 1365/4393 [1:57:29<4:53:41, 5.82s/it] + Training...: 31% 1366/4393 [1:57:35<4:51:31, 5.78s/it] + Training...: 31% 1367/4393 [1:57:40<4:48:25, 5.72s/it] + Training...: 31% 1368/4393 [1:57:46<4:44:51, 5.65s/it] + Training...: 31% 1369/4393 [1:57:51<4:42:55, 5.61s/it] + Training...: 31% 1370/4393 [1:57:57<4:41:25, 5.59s/it] + Training...: 31% 1371/4393 [1:58:02<4:39:08, 5.54s/it] + Training...: 31% 1372/4393 [1:58:08<4:35:45, 5.48s/it] + Training...: 31% 1373/4393 [1:58:13<4:35:05, 5.47s/it] + Training...: 31% 1374/4393 [1:58:19<4:34:23, 5.45s/it] + Training...: 31% 1375/4393 [1:58:24<4:31:34, 5.40s/it] + Training...: 31% 1376/4393 [1:58:29<4:28:02, 5.33s/it] + Training...: 31% 1377/4393 [1:58:34<4:24:20, 5.26s/it] + Training...: 31% 1378/4393 [1:58:39<4:21:01, 5.19s/it] + Training...: 31% 1379/4393 [1:58:44<4:18:41, 5.15s/it] + Training...: 31% 1380/4393 [1:58:49<4:15:16, 5.08s/it] + Training...: 31% 1381/4393 [1:58:54<4:13:12, 5.04s/it] + Training...: 31% 1382/4393 [1:58:59<4:08:53, 4.96s/it] + Training...: 31% 1383/4393 [1:59:04<4:06:51, 4.92s/it] + Training...: 32% 1384/4393 [1:59:08<4:03:28, 4.86s/it] + Training...: 32% 1385/4393 [1:59:13<4:00:42, 4.80s/it] + Training...: 32% 1386/4393 [1:59:18<3:56:51, 4.73s/it] + Training...: 32% 1387/4393 [1:59:22<3:51:41, 4.62s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:36:27<17:31:17, 21025.84s/it] + Training...: 32% 1387/4393 [1:59:27<3:51:41, 4.62s/it] + Training...: 32% 1388/4393 [1:59:27<3:53:57, 4.67s/it] + Training...: 32% 1389/4393 [1:59:31<3:42:23, 4.44s/it] + Training...: 32% 1390/4393 [1:59:34<3:30:20, 4.20s/it] + Training...: 32% 1391/4393 [1:59:38<3:19:03, 3.98s/it] + Training...: 32% 1392/4393 [1:59:41<3:07:15, 3.74s/it] + Training...: 32% 1393/4393 [1:59:44<2:54:59, 3.50s/it] + Training...: 32% 1394/4393 [1:59:47<2:43:15, 3.27s/it] + Training...: 32% 1395/4393 [1:59:49<2:30:10, 3.01s/it] + Training...: 32% 1396/4393 [1:59:51<2:16:11, 2.73s/it] + Training...: 32% 1397/4393 [1:59:53<2:01:45, 2.44s/it] + Training...: 32% 1398/4393 [1:59:54<1:47:48, 2.16s/it] + Training...: 32% 1399/4393 [1:59:56<1:34:45, 1.90s/it] + Training...: 32% 1400/4393 [1:59:57<1:21:49, 1.64s/it] + Training...: 32% 1401/4393 [2:00:03<2:31:21, 3.04s/it] + Training...: 32% 1402/4393 [2:00:09<3:19:49, 4.01s/it] + Training...: 32% 1403/4393 [2:00:15<3:51:08, 4.64s/it] + Training...: 32% 1404/4393 [2:00:21<4:10:52, 5.04s/it] + Training...: 32% 1405/4393 [2:00:27<4:25:26, 5.33s/it] + Training...: 32% 1406/4393 [2:00:33<4:34:36, 5.52s/it] + Training...: 32% 1407/4393 [2:00:39<4:39:54, 5.62s/it] + Training...: 32% 1408/4393 [2:00:45<4:42:40, 5.68s/it] + Training...: 32% 1409/4393 [2:00:51<4:45:03, 5.73s/it] + Training...: 32% 1410/4393 [2:00:57<4:46:19, 5.76s/it] + Training...: 32% 1411/4393 [2:01:03<4:47:51, 5.79s/it] + Training...: 32% 1412/4393 [2:01:08<4:46:51, 5.77s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:38:15<17:31:17, 21025.84s/it] + Training...: 32% 1412/4393 [2:01:15<4:46:51, 5.77s/it] + Training...: 32% 1413/4393 [2:01:15<4:57:43, 5.99s/it] + Training...: 32% 1414/4393 [2:01:20<4:52:48, 5.90s/it] + Training...: 32% 1415/4393 [2:01:26<4:48:28, 5.81s/it] + Training...: 32% 1416/4393 [2:01:32<4:48:08, 5.81s/it] + Training...: 32% 1417/4393 [2:01:38<4:47:42, 5.80s/it] + Training...: 32% 1418/4393 [2:01:43<4:45:27, 5.76s/it] + Training...: 32% 1419/4393 [2:01:49<4:44:21, 5.74s/it] + Training...: 32% 1420/4393 [2:01:54<4:39:19, 5.64s/it] + Training...: 32% 1421/4393 [2:02:00<4:35:33, 5.56s/it] + Training...: 32% 1422/4393 [2:02:05<4:32:27, 5.50s/it] + Training...: 32% 1423/4393 [2:02:10<4:30:07, 5.46s/it] + Training...: 32% 1424/4393 [2:02:16<4:27:31, 5.41s/it] + Training...: 32% 1425/4393 [2:02:21<4:25:56, 5.38s/it] + Training...: 32% 1426/4393 [2:02:26<4:23:10, 5.32s/it] + Training...: 32% 1427/4393 [2:02:31<4:21:21, 5.29s/it] + Training...: 33% 1428/4393 [2:02:37<4:18:09, 5.22s/it] + Training...: 33% 1429/4393 [2:02:42<4:16:12, 5.19s/it] + Training...: 33% 1430/4393 [2:02:47<4:12:30, 5.11s/it] + Training...: 33% 1431/4393 [2:02:51<4:09:08, 5.05s/it] + Training...: 33% 1432/4393 [2:02:56<4:05:41, 4.98s/it] + Training...: 33% 1433/4393 [2:03:01<4:02:29, 4.92s/it] + Training...: 33% 1434/4393 [2:03:06<3:58:20, 4.83s/it] + Training...: 33% 1435/4393 [2:03:10<3:53:16, 4.73s/it] + Training...: 33% 1436/4393 [2:03:15<3:47:33, 4.62s/it] + Training...: 33% 1437/4393 [2:03:19<3:42:07, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:40:23<17:31:17, 21025.84s/it] + Training...: 33% 1437/4393 [2:03:23<3:42:07, 4.51s/it] + Training...: 33% 1438/4393 [2:03:23<3:44:19, 4.55s/it] + Training...: 33% 1439/4393 [2:03:27<3:34:16, 4.35s/it] + Training...: 33% 1440/4393 [2:03:31<3:23:02, 4.13s/it] + Training...: 33% 1441/4393 [2:03:34<3:11:49, 3.90s/it] + Training...: 33% 1442/4393 [2:03:37<3:00:33, 3.67s/it] + Training...: 33% 1443/4393 [2:03:40<2:48:39, 3.43s/it] + Training...: 33% 1444/4393 [2:03:43<2:36:04, 3.18s/it] + Training...: 33% 1445/4393 [2:03:45<2:22:43, 2.90s/it] + Training...: 33% 1446/4393 [2:03:47<2:09:06, 2.63s/it] + Training...: 33% 1447/4393 [2:03:49<1:55:49, 2.36s/it] + Training...: 33% 1448/4393 [2:03:50<1:43:31, 2.11s/it] + Training...: 33% 1449/4393 [2:03:52<1:31:09, 1.86s/it] + Training...: 33% 1450/4393 [2:03:53<1:18:19, 1.60s/it] + Training...: 33% 1451/4393 [2:03:59<2:25:52, 2.98s/it] + Training...: 33% 1452/4393 [2:04:05<3:13:58, 3.96s/it] + Training...: 33% 1453/4393 [2:04:11<3:44:26, 4.58s/it] + Training...: 33% 1454/4393 [2:04:17<4:05:06, 5.00s/it] + Training...: 33% 1455/4393 [2:04:23<4:21:05, 5.33s/it] + Training...: 33% 1456/4393 [2:04:29<4:30:38, 5.53s/it] + Training...: 33% 1457/4393 [2:04:35<4:37:59, 5.68s/it] + Training...: 33% 1458/4393 [2:04:41<4:40:13, 5.73s/it] + Training...: 33% 1459/4393 [2:04:47<4:42:45, 5.78s/it] + Training...: 33% 1460/4393 [2:04:53<4:42:10, 5.77s/it] + Training...: 33% 1461/4393 [2:04:59<4:42:55, 5.79s/it] + Training...: 33% 1462/4393 [2:05:04<4:42:46, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:42:11<17:31:17, 21025.84s/it] + Training...: 33% 1462/4393 [2:05:11<4:42:46, 5.79s/it] + Training...: 33% 1463/4393 [2:05:11<4:53:36, 6.01s/it] + Training...: 33% 1464/4393 [2:05:17<4:48:19, 5.91s/it] + Training...: 33% 1465/4393 [2:05:22<4:44:37, 5.83s/it] + Training...: 33% 1466/4393 [2:05:28<4:41:14, 5.76s/it] + Training...: 33% 1467/4393 [2:05:34<4:40:07, 5.74s/it] + Training...: 33% 1468/4393 [2:05:39<4:36:23, 5.67s/it] + Training...: 33% 1469/4393 [2:05:45<4:34:25, 5.63s/it] + Training...: 33% 1470/4393 [2:05:50<4:31:22, 5.57s/it] + Training...: 33% 1471/4393 [2:05:56<4:30:40, 5.56s/it] + Training...: 34% 1472/4393 [2:06:01<4:30:01, 5.55s/it] + Training...: 34% 1473/4393 [2:06:06<4:27:40, 5.50s/it] + Training...: 34% 1474/4393 [2:06:12<4:24:13, 5.43s/it] + Training...: 34% 1475/4393 [2:06:17<4:21:49, 5.38s/it] + Training...: 34% 1476/4393 [2:06:22<4:20:24, 5.36s/it] + Training...: 34% 1477/4393 [2:06:27<4:18:00, 5.31s/it] + Training...: 34% 1478/4393 [2:06:33<4:15:05, 5.25s/it] + Training...: 34% 1479/4393 [2:06:38<4:11:49, 5.19s/it] + Training...: 34% 1480/4393 [2:06:43<4:08:50, 5.13s/it] + Training...: 34% 1481/4393 [2:06:48<4:07:01, 5.09s/it] + Training...: 34% 1482/4393 [2:06:52<4:03:36, 5.02s/it] + Training...: 34% 1483/4393 [2:06:57<4:00:31, 4.96s/it] + Training...: 34% 1484/4393 [2:07:02<3:55:33, 4.86s/it] + Training...: 34% 1485/4393 [2:07:06<3:51:28, 4.78s/it] + Training...: 34% 1486/4393 [2:07:11<3:46:01, 4.67s/it] + Training...: 34% 1487/4393 [2:07:15<3:41:55, 4.58s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:44:20<17:31:17, 21025.84s/it] + Training...: 34% 1487/4393 [2:07:20<3:41:55, 4.58s/it] + Training...: 34% 1488/4393 [2:07:20<3:44:39, 4.64s/it] + Training...: 34% 1489/4393 [2:07:24<3:34:21, 4.43s/it] + Training...: 34% 1490/4393 [2:07:28<3:23:34, 4.21s/it] + Training...: 34% 1491/4393 [2:07:31<3:13:15, 4.00s/it] + Training...: 34% 1492/4393 [2:07:34<3:02:27, 3.77s/it] + Training...: 34% 1493/4393 [2:07:37<2:50:47, 3.53s/it] + Training...: 34% 1494/4393 [2:07:40<2:38:53, 3.29s/it] + Training...: 34% 1495/4393 [2:07:43<2:26:13, 3.03s/it] + Training...: 34% 1496/4393 [2:07:45<2:13:09, 2.76s/it] + Training...: 34% 1497/4393 [2:07:47<1:59:42, 2.48s/it] + Training...: 34% 1498/4393 [2:07:48<1:46:25, 2.21s/it] + Training...: 34% 1499/4393 [2:07:49<1:32:32, 1.92s/it] + Training...: 34% 1500/4393 [2:07:50<1:18:57, 1.64s/it] + Training...: 34% 1501/4393 [2:07:57<2:25:50, 3.03s/it] + Training...: 34% 1502/4393 [2:08:03<3:12:07, 3.99s/it] + Training...: 34% 1503/4393 [2:08:09<3:43:05, 4.63s/it] + Training...: 34% 1504/4393 [2:08:15<4:02:25, 5.03s/it] + Training...: 34% 1505/4393 [2:08:21<4:16:35, 5.33s/it] + Training...: 34% 1506/4393 [2:08:27<4:24:43, 5.50s/it] + Training...: 34% 1507/4393 [2:08:33<4:30:10, 5.62s/it] + Training...: 34% 1508/4393 [2:08:39<4:33:03, 5.68s/it] + Training...: 34% 1509/4393 [2:08:44<4:35:41, 5.74s/it] + Training...: 34% 1510/4393 [2:08:50<4:35:55, 5.74s/it] + Training...: 34% 1511/4393 [2:08:56<4:38:24, 5.80s/it] + Training...: 34% 1512/4393 [2:09:02<4:40:07, 5.83s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:46:08<17:31:17, 21025.84s/it] + Training...: 34% 1512/4393 [2:09:09<4:40:07, 5.83s/it] + Training...: 34% 1513/4393 [2:09:09<4:49:39, 6.03s/it] + Training...: 34% 1514/4393 [2:09:14<4:43:41, 5.91s/it] + Training...: 34% 1515/4393 [2:09:20<4:39:19, 5.82s/it] + Training...: 35% 1516/4393 [2:09:25<4:35:31, 5.75s/it] + Training...: 35% 1517/4393 [2:09:31<4:33:36, 5.71s/it] + Training...: 35% 1518/4393 [2:09:37<4:31:56, 5.68s/it] + Training...: 35% 1519/4393 [2:09:42<4:29:41, 5.63s/it] + Training...: 35% 1520/4393 [2:09:48<4:26:45, 5.57s/it] + Training...: 35% 1521/4393 [2:09:53<4:24:16, 5.52s/it] + Training...: 35% 1522/4393 [2:09:58<4:22:01, 5.48s/it] + Training...: 35% 1523/4393 [2:10:04<4:19:55, 5.43s/it] + Training...: 35% 1524/4393 [2:10:09<4:17:51, 5.39s/it] + Training...: 35% 1525/4393 [2:10:14<4:16:09, 5.36s/it] + Training...: 35% 1526/4393 [2:10:19<4:13:49, 5.31s/it] + Training...: 35% 1527/4393 [2:10:25<4:12:48, 5.29s/it] + Training...: 35% 1528/4393 [2:10:30<4:10:05, 5.24s/it] + Training...: 35% 1529/4393 [2:10:35<4:07:01, 5.18s/it] + Training...: 35% 1530/4393 [2:10:40<4:04:11, 5.12s/it] + Training...: 35% 1531/4393 [2:10:45<4:00:41, 5.05s/it] + Training...: 35% 1532/4393 [2:10:49<3:57:16, 4.98s/it] + Training...: 35% 1533/4393 [2:10:54<3:53:46, 4.90s/it] + Training...: 35% 1534/4393 [2:10:59<3:49:57, 4.83s/it] + Training...: 35% 1535/4393 [2:11:03<3:47:00, 4.77s/it] + Training...: 35% 1536/4393 [2:11:08<3:43:37, 4.70s/it] + Training...: 35% 1537/4393 [2:11:12<3:38:25, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:48:17<17:31:17, 21025.84s/it] + Training...: 35% 1537/4393 [2:11:17<3:38:25, 4.59s/it] + Training...: 35% 1538/4393 [2:11:17<3:41:04, 4.65s/it] + Training...: 35% 1539/4393 [2:11:21<3:32:16, 4.46s/it] + Training...: 35% 1540/4393 [2:11:25<3:22:53, 4.27s/it] + Training...: 35% 1541/4393 [2:11:29<3:13:35, 4.07s/it] + Training...: 35% 1542/4393 [2:11:32<3:02:19, 3.84s/it] + Training...: 35% 1543/4393 [2:11:35<2:50:31, 3.59s/it] + Training...: 35% 1544/4393 [2:11:38<2:37:49, 3.32s/it] + Training...: 35% 1545/4393 [2:11:40<2:24:59, 3.05s/it] + Training...: 35% 1546/4393 [2:11:42<2:11:41, 2.78s/it] + Training...: 35% 1547/4393 [2:11:44<1:58:36, 2.50s/it] + Training...: 35% 1548/4393 [2:11:46<1:45:33, 2.23s/it] + Training...: 35% 1549/4393 [2:11:47<1:32:29, 1.95s/it] + Training...: 35% 1550/4393 [2:11:48<1:19:27, 1.68s/it] + Training...: 35% 1551/4393 [2:11:54<2:23:30, 3.03s/it] + Training...: 35% 1552/4393 [2:12:00<3:09:39, 4.01s/it] + Training...: 35% 1553/4393 [2:12:06<3:38:27, 4.62s/it] + Training...: 35% 1554/4393 [2:12:12<3:57:20, 5.02s/it] + Training...: 35% 1555/4393 [2:12:18<4:10:52, 5.30s/it] + Training...: 35% 1556/4393 [2:12:24<4:19:42, 5.49s/it] + Training...: 35% 1557/4393 [2:12:30<4:25:56, 5.63s/it] + Training...: 35% 1558/4393 [2:12:36<4:28:52, 5.69s/it] + Training...: 35% 1559/4393 [2:12:42<4:31:33, 5.75s/it] + Training...: 36% 1560/4393 [2:12:48<4:31:36, 5.75s/it] + Training...: 36% 1561/4393 [2:12:54<4:32:08, 5.77s/it] + Training...: 36% 1562/4393 [2:13:00<4:38:17, 5.90s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:50:07<17:31:17, 21025.84s/it] + Training...: 36% 1562/4393 [2:13:07<4:38:17, 5.90s/it] + Training...: 36% 1563/4393 [2:13:07<4:54:07, 6.24s/it] + Training...: 36% 1564/4393 [2:13:13<4:47:09, 6.09s/it] + Training...: 36% 1565/4393 [2:13:18<4:41:02, 5.96s/it] + Training...: 36% 1566/4393 [2:13:24<4:35:13, 5.84s/it] + Training...: 36% 1567/4393 [2:13:29<4:31:31, 5.76s/it] + Training...: 36% 1568/4393 [2:13:35<4:28:02, 5.69s/it] + Training...: 36% 1569/4393 [2:13:40<4:26:38, 5.67s/it] + Training...: 36% 1570/4393 [2:13:46<4:25:26, 5.64s/it] + Training...: 36% 1571/4393 [2:13:52<4:26:23, 5.66s/it] + Training...: 36% 1572/4393 [2:13:57<4:23:44, 5.61s/it] + Training...: 36% 1573/4393 [2:14:03<4:22:09, 5.58s/it] + Training...: 36% 1574/4393 [2:14:08<4:20:13, 5.54s/it] + Training...: 36% 1575/4393 [2:14:14<4:18:44, 5.51s/it] + Training...: 36% 1576/4393 [2:14:19<4:16:37, 5.47s/it] + Training...: 36% 1577/4393 [2:14:24<4:15:14, 5.44s/it] + Training...: 36% 1578/4393 [2:14:30<4:16:29, 5.47s/it] + Training...: 36% 1579/4393 [2:14:35<4:14:14, 5.42s/it] + Training...: 36% 1580/4393 [2:14:40<4:10:17, 5.34s/it] + Training...: 36% 1581/4393 [2:14:45<4:06:40, 5.26s/it] + Training...: 36% 1582/4393 [2:14:50<4:02:06, 5.17s/it] + Training...: 36% 1583/4393 [2:14:55<3:57:33, 5.07s/it] + Training...: 36% 1584/4393 [2:15:00<3:52:07, 4.96s/it] + Training...: 36% 1585/4393 [2:15:05<3:48:02, 4.87s/it] + Training...: 36% 1586/4393 [2:15:09<3:42:42, 4.76s/it] + Training...: 36% 1587/4393 [2:15:13<3:36:57, 4.64s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:52:18<17:31:17, 21025.84s/it] + Training...: 36% 1587/4393 [2:15:18<3:36:57, 4.64s/it] + Training...: 36% 1588/4393 [2:15:18<3:38:40, 4.68s/it] + Training...: 36% 1589/4393 [2:15:22<3:26:19, 4.41s/it] + Training...: 36% 1590/4393 [2:15:26<3:13:54, 4.15s/it] + Training...: 36% 1591/4393 [2:15:29<3:02:12, 3.90s/it] + Training...: 36% 1592/4393 [2:15:32<2:49:29, 3.63s/it] + Training...: 36% 1593/4393 [2:15:35<2:37:37, 3.38s/it] + Training...: 36% 1594/4393 [2:15:37<2:26:20, 3.14s/it] + Training...: 36% 1595/4393 [2:15:40<2:14:15, 2.88s/it] + Training...: 36% 1596/4393 [2:15:42<2:02:36, 2.63s/it] + Training...: 36% 1597/4393 [2:15:43<1:50:27, 2.37s/it] + Training...: 36% 1598/4393 [2:15:45<1:39:31, 2.14s/it] + Training...: 36% 1599/4393 [2:15:46<1:27:40, 1.88s/it] + Training...: 36% 1600/4393 [2:15:47<1:15:53, 1.63s/it] + Training...: 36% 1601/4393 [2:15:54<2:22:57, 3.07s/it] + Training...: 36% 1602/4393 [2:16:00<3:07:19, 4.03s/it] + Training...: 36% 1603/4393 [2:16:06<3:35:40, 4.64s/it] + Training...: 37% 1604/4393 [2:16:12<3:54:40, 5.05s/it] + Training...: 37% 1605/4393 [2:16:18<4:08:12, 5.34s/it] + Training...: 37% 1606/4393 [2:16:24<4:16:39, 5.53s/it] + Training...: 37% 1607/4393 [2:16:30<4:22:03, 5.64s/it] + Training...: 37% 1608/4393 [2:16:36<4:24:26, 5.70s/it] + Training...: 37% 1609/4393 [2:16:42<4:26:30, 5.74s/it] + Training...: 37% 1610/4393 [2:16:47<4:26:32, 5.75s/it] + Training...: 37% 1611/4393 [2:16:53<4:26:49, 5.75s/it] + Training...: 37% 1612/4393 [2:16:59<4:27:03, 5.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:54:05<17:31:17, 21025.84s/it] + Training...: 37% 1612/4393 [2:17:06<4:27:03, 5.76s/it] + Training...: 37% 1613/4393 [2:17:06<4:39:07, 6.02s/it] + Training...: 37% 1614/4393 [2:17:11<4:37:49, 6.00s/it] + Training...: 37% 1615/4393 [2:17:17<4:33:59, 5.92s/it] + Training...: 37% 1616/4393 [2:17:23<4:29:17, 5.82s/it] + Training...: 37% 1617/4393 [2:17:28<4:25:41, 5.74s/it] + Training...: 37% 1618/4393 [2:17:34<4:22:19, 5.67s/it] + Training...: 37% 1619/4393 [2:17:39<4:19:54, 5.62s/it] + Training...: 37% 1620/4393 [2:17:45<4:18:23, 5.59s/it] + Training...: 37% 1621/4393 [2:17:50<4:16:29, 5.55s/it] + Training...: 37% 1622/4393 [2:17:56<4:14:19, 5.51s/it] + Training...: 37% 1623/4393 [2:18:01<4:12:53, 5.48s/it] + Training...: 37% 1624/4393 [2:18:06<4:10:22, 5.43s/it] + Training...: 37% 1625/4393 [2:18:12<4:09:22, 5.41s/it] + Training...: 37% 1626/4393 [2:18:17<4:06:13, 5.34s/it] + Training...: 37% 1627/4393 [2:18:22<4:04:23, 5.30s/it] + Training...: 37% 1628/4393 [2:18:27<4:01:22, 5.24s/it] + Training...: 37% 1629/4393 [2:18:32<3:59:25, 5.20s/it] + Training...: 37% 1630/4393 [2:18:37<3:56:55, 5.14s/it] + Training...: 37% 1631/4393 [2:18:42<3:54:00, 5.08s/it] + Training...: 37% 1632/4393 [2:18:47<3:53:04, 5.06s/it] + Training...: 37% 1633/4393 [2:18:52<3:51:14, 5.03s/it] + Training...: 37% 1634/4393 [2:18:57<3:46:06, 4.92s/it] + Training...: 37% 1635/4393 [2:19:02<3:41:55, 4.83s/it] + Training...: 37% 1636/4393 [2:19:06<3:37:07, 4.73s/it] + Training...: 37% 1637/4393 [2:19:10<3:31:02, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:56:15<17:31:17, 21025.84s/it] + Training...: 37% 1637/4393 [2:19:15<3:31:02, 4.59s/it] + Training...: 37% 1638/4393 [2:19:15<3:32:55, 4.64s/it] + Training...: 37% 1639/4393 [2:19:19<3:23:01, 4.42s/it] + Training...: 37% 1640/4393 [2:19:23<3:12:18, 4.19s/it] + Training...: 37% 1641/4393 [2:19:26<3:01:01, 3.95s/it] + Training...: 37% 1642/4393 [2:19:29<2:49:28, 3.70s/it] + Training...: 37% 1643/4393 [2:19:32<2:38:03, 3.45s/it] + Training...: 37% 1644/4393 [2:19:35<2:26:04, 3.19s/it] + Training...: 37% 1645/4393 [2:19:37<2:12:51, 2.90s/it] + Training...: 37% 1646/4393 [2:19:39<2:00:59, 2.64s/it] + Training...: 37% 1647/4393 [2:19:41<1:48:57, 2.38s/it] + Training...: 38% 1648/4393 [2:19:42<1:37:28, 2.13s/it] + Training...: 38% 1649/4393 [2:19:44<1:26:09, 1.88s/it] + Training...: 38% 1650/4393 [2:19:45<1:14:49, 1.64s/it] + Training...: 38% 1651/4393 [2:19:51<2:19:15, 3.05s/it] + Training...: 38% 1652/4393 [2:19:57<3:04:00, 4.03s/it] + Training...: 38% 1653/4393 [2:20:03<3:32:05, 4.64s/it] + Training...: 38% 1654/4393 [2:20:09<3:50:24, 5.05s/it] + Training...: 38% 1655/4393 [2:20:15<4:03:49, 5.34s/it] + Training...: 38% 1656/4393 [2:20:21<4:11:40, 5.52s/it] + Training...: 38% 1657/4393 [2:20:27<4:16:37, 5.63s/it] + Training...: 38% 1658/4393 [2:20:33<4:19:06, 5.68s/it] + Training...: 38% 1659/4393 [2:20:39<4:21:41, 5.74s/it] + Training...: 38% 1660/4393 [2:20:45<4:21:32, 5.74s/it] + Training...: 38% 1661/4393 [2:20:50<4:22:42, 5.77s/it] + Training...: 38% 1662/4393 [2:20:56<4:24:13, 5.81s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [54:58:03<17:31:17, 21025.84s/it] + Training...: 38% 1662/4393 [2:21:03<4:24:13, 5.81s/it] + Training...: 38% 1663/4393 [2:21:03<4:35:41, 6.06s/it] + Training...: 38% 1664/4393 [2:21:09<4:32:06, 5.98s/it] + Training...: 38% 1665/4393 [2:21:15<4:29:12, 5.92s/it] + Training...: 38% 1666/4393 [2:21:20<4:24:13, 5.81s/it] + Training...: 38% 1667/4393 [2:21:26<4:20:56, 5.74s/it] + Training...: 38% 1668/4393 [2:21:31<4:17:00, 5.66s/it] + Training...: 38% 1669/4393 [2:21:37<4:15:37, 5.63s/it] + Training...: 38% 1670/4393 [2:21:42<4:13:39, 5.59s/it] + Training...: 38% 1671/4393 [2:21:48<4:11:26, 5.54s/it] + Training...: 38% 1672/4393 [2:21:53<4:09:14, 5.50s/it] + Training...: 38% 1673/4393 [2:21:58<4:07:49, 5.47s/it] + Training...: 38% 1674/4393 [2:22:04<4:05:21, 5.41s/it] + Training...: 38% 1675/4393 [2:22:09<4:03:15, 5.37s/it] + Training...: 38% 1676/4393 [2:22:14<4:01:44, 5.34s/it] + Training...: 38% 1677/4393 [2:22:19<3:59:32, 5.29s/it] + Training...: 38% 1678/4393 [2:22:25<3:58:01, 5.26s/it] + Training...: 38% 1679/4393 [2:22:30<3:56:46, 5.23s/it] + Training...: 38% 1680/4393 [2:22:35<3:56:47, 5.24s/it] + Training...: 38% 1681/4393 [2:22:40<3:52:51, 5.15s/it] + Training...: 38% 1682/4393 [2:22:45<3:47:31, 5.04s/it] + Training...: 38% 1683/4393 [2:22:50<3:44:33, 4.97s/it] + Training...: 38% 1684/4393 [2:22:54<3:40:48, 4.89s/it] + Training...: 38% 1685/4393 [2:22:59<3:36:24, 4.79s/it] + Training...: 38% 1686/4393 [2:23:03<3:30:30, 4.67s/it] + Training...: 38% 1687/4393 [2:23:07<3:24:20, 4.53s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:00:12<17:31:17, 21025.84s/it] + Training...: 38% 1687/4393 [2:23:12<3:24:20, 4.53s/it] + Training...: 38% 1688/4393 [2:23:12<3:25:50, 4.57s/it] + Training...: 38% 1689/4393 [2:23:16<3:15:54, 4.35s/it] + Training...: 38% 1690/4393 [2:23:19<3:05:17, 4.11s/it] + Training...: 38% 1691/4393 [2:23:23<2:54:08, 3.87s/it] + Training...: 39% 1692/4393 [2:23:26<2:42:45, 3.62s/it] + Training...: 39% 1693/4393 [2:23:29<2:31:00, 3.36s/it] + Training...: 39% 1694/4393 [2:23:31<2:19:00, 3.09s/it] + Training...: 39% 1695/4393 [2:23:33<2:07:11, 2.83s/it] + Training...: 39% 1696/4393 [2:23:35<1:55:23, 2.57s/it] + Training...: 39% 1697/4393 [2:23:37<1:43:23, 2.30s/it] + Training...: 39% 1698/4393 [2:23:38<1:32:18, 2.06s/it] + Training...: 39% 1699/4393 [2:23:40<1:21:25, 1.81s/it] + Training...: 39% 1700/4393 [2:23:41<1:10:25, 1.57s/it] + Training...: 39% 1701/4393 [2:23:47<2:14:03, 2.99s/it] + Training...: 39% 1702/4393 [2:23:53<2:58:25, 3.98s/it] + Training...: 39% 1703/4393 [2:23:59<3:26:50, 4.61s/it] + Training...: 39% 1704/4393 [2:24:05<3:45:55, 5.04s/it] + Training...: 39% 1705/4393 [2:24:11<3:58:40, 5.33s/it] + Training...: 39% 1706/4393 [2:24:17<4:06:35, 5.51s/it] + Training...: 39% 1707/4393 [2:24:23<4:12:31, 5.64s/it] + Training...: 39% 1708/4393 [2:24:29<4:15:33, 5.71s/it] + Training...: 39% 1709/4393 [2:24:35<4:17:01, 5.75s/it] + Training...: 39% 1710/4393 [2:24:41<4:17:04, 5.75s/it] + Training...: 39% 1711/4393 [2:24:46<4:17:53, 5.77s/it] + Training...: 39% 1712/4393 [2:24:52<4:19:29, 5.81s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:01:59<17:31:17, 21025.84s/it] + Training...: 39% 1712/4393 [2:24:59<4:19:29, 5.81s/it] + Training...: 39% 1713/4393 [2:24:59<4:30:43, 6.06s/it] + Training...: 39% 1714/4393 [2:25:05<4:26:17, 5.96s/it] + Training...: 39% 1715/4393 [2:25:11<4:23:54, 5.91s/it] + Training...: 39% 1716/4393 [2:25:16<4:20:07, 5.83s/it] + Training...: 39% 1717/4393 [2:25:22<4:17:01, 5.76s/it] + Training...: 39% 1718/4393 [2:25:27<4:13:16, 5.68s/it] + Training...: 39% 1719/4393 [2:25:33<4:11:28, 5.64s/it] + Training...: 39% 1720/4393 [2:25:38<4:09:19, 5.60s/it] + Training...: 39% 1721/4393 [2:25:44<4:06:23, 5.53s/it] + Training...: 39% 1722/4393 [2:25:49<4:03:53, 5.48s/it] + Training...: 39% 1723/4393 [2:25:54<4:02:41, 5.45s/it] + Training...: 39% 1724/4393 [2:26:00<4:00:46, 5.41s/it] + Training...: 39% 1725/4393 [2:26:05<3:59:13, 5.38s/it] + Training...: 39% 1726/4393 [2:26:10<3:56:22, 5.32s/it] + Training...: 39% 1727/4393 [2:26:15<3:54:09, 5.27s/it] + Training...: 39% 1728/4393 [2:26:20<3:51:17, 5.21s/it] + Training...: 39% 1729/4393 [2:26:26<3:48:59, 5.16s/it] + Training...: 39% 1730/4393 [2:26:30<3:46:07, 5.09s/it] + Training...: 39% 1731/4393 [2:26:35<3:43:46, 5.04s/it] + Training...: 39% 1732/4393 [2:26:40<3:39:52, 4.96s/it] + Training...: 39% 1733/4393 [2:26:45<3:35:31, 4.86s/it] + Training...: 39% 1734/4393 [2:26:49<3:31:34, 4.77s/it] + Training...: 39% 1735/4393 [2:26:54<3:26:32, 4.66s/it] + Training...: 40% 1736/4393 [2:26:58<3:20:12, 4.52s/it] + Training...: 40% 1737/4393 [2:27:02<3:15:17, 4.41s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:04:07<17:31:17, 21025.84s/it] + Training...: 40% 1737/4393 [2:27:07<3:15:17, 4.41s/it] + Training...: 40% 1738/4393 [2:27:07<3:16:44, 4.45s/it] + Training...: 40% 1739/4393 [2:27:10<3:06:15, 4.21s/it] + Training...: 40% 1740/4393 [2:27:14<2:55:14, 3.96s/it] + Training...: 40% 1741/4393 [2:27:17<2:43:34, 3.70s/it] + Training...: 40% 1742/4393 [2:27:20<2:32:07, 3.44s/it] + Training...: 40% 1743/4393 [2:27:22<2:20:34, 3.18s/it] + Training...: 40% 1744/4393 [2:27:25<2:09:21, 2.93s/it] + Training...: 40% 1745/4393 [2:27:27<1:58:25, 2.68s/it] + Training...: 40% 1746/4393 [2:27:29<1:47:55, 2.45s/it] + Training...: 40% 1747/4393 [2:27:30<1:37:24, 2.21s/it] + Training...: 40% 1748/4393 [2:27:32<1:26:59, 1.97s/it] + Training...: 40% 1749/4393 [2:27:33<1:16:43, 1.74s/it] + Training...: 40% 1750/4393 [2:27:34<1:06:36, 1.51s/it] + Training...: 40% 1751/4393 [2:27:40<2:08:20, 2.91s/it] + Training...: 40% 1752/4393 [2:27:46<2:52:01, 3.91s/it] + Training...: 40% 1753/4393 [2:27:52<3:21:17, 4.57s/it] + Training...: 40% 1754/4393 [2:27:58<3:39:23, 4.99s/it] + Training...: 40% 1755/4393 [2:28:04<3:53:11, 5.30s/it] + Training...: 40% 1756/4393 [2:28:10<4:00:42, 5.48s/it] + Training...: 40% 1757/4393 [2:28:16<4:06:30, 5.61s/it] + Training...: 40% 1758/4393 [2:28:22<4:09:21, 5.68s/it] + Training...: 40% 1759/4393 [2:28:28<4:12:20, 5.75s/it] + Training...: 40% 1760/4393 [2:28:34<4:13:17, 5.77s/it] + Training...: 40% 1761/4393 [2:28:40<4:13:42, 5.78s/it] + Training...: 40% 1762/4393 [2:28:45<4:16:06, 5.84s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:05:52<17:31:17, 21025.84s/it] + Training...: 40% 1762/4393 [2:28:52<4:16:06, 5.84s/it] + Training...: 40% 1763/4393 [2:28:52<4:25:13, 6.05s/it] + Training...: 40% 1764/4393 [2:28:58<4:19:32, 5.92s/it] + Training...: 40% 1765/4393 [2:29:03<4:16:17, 5.85s/it] + Training...: 40% 1766/4393 [2:29:09<4:13:00, 5.78s/it] + Training...: 40% 1767/4393 [2:29:15<4:10:55, 5.73s/it] + Training...: 40% 1768/4393 [2:29:20<4:07:59, 5.67s/it] + Training...: 40% 1769/4393 [2:29:26<4:05:31, 5.61s/it] + Training...: 40% 1770/4393 [2:29:31<4:02:45, 5.55s/it] + Training...: 40% 1771/4393 [2:29:36<4:00:49, 5.51s/it] + Training...: 40% 1772/4393 [2:29:42<3:58:24, 5.46s/it] + Training...: 40% 1773/4393 [2:29:47<3:56:42, 5.42s/it] + Training...: 40% 1774/4393 [2:29:52<3:54:30, 5.37s/it] + Training...: 40% 1775/4393 [2:29:58<3:52:55, 5.34s/it] + Training...: 40% 1776/4393 [2:30:03<3:50:30, 5.29s/it] + Training...: 40% 1777/4393 [2:30:08<3:51:19, 5.31s/it] + Training...: 40% 1778/4393 [2:30:13<3:50:04, 5.28s/it] + Training...: 40% 1779/4393 [2:30:18<3:47:09, 5.21s/it] + Training...: 41% 1780/4393 [2:30:23<3:43:34, 5.13s/it] + Training...: 41% 1781/4393 [2:30:28<3:40:25, 5.06s/it] + Training...: 41% 1782/4393 [2:30:33<3:36:17, 4.97s/it] + Training...: 41% 1783/4393 [2:30:38<3:32:40, 4.89s/it] + Training...: 41% 1784/4393 [2:30:42<3:28:33, 4.80s/it] + Training...: 41% 1785/4393 [2:30:47<3:25:10, 4.72s/it] + Training...: 41% 1786/4393 [2:30:51<3:19:35, 4.59s/it] + Training...: 41% 1787/4393 [2:30:55<3:14:53, 4.49s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:08:00<17:31:17, 21025.84s/it] + Training...: 41% 1787/4393 [2:31:00<3:14:53, 4.49s/it] + Training...: 41% 1788/4393 [2:31:00<3:17:13, 4.54s/it] + Training...: 41% 1789/4393 [2:31:04<3:09:00, 4.35s/it] + Training...: 41% 1790/4393 [2:31:07<2:58:32, 4.12s/it] + Training...: 41% 1791/4393 [2:31:11<2:48:16, 3.88s/it] + Training...: 41% 1792/4393 [2:31:14<2:37:26, 3.63s/it] + Training...: 41% 1793/4393 [2:31:17<2:26:12, 3.37s/it] + Training...: 41% 1794/4393 [2:31:19<2:14:46, 3.11s/it] + Training...: 41% 1795/4393 [2:31:21<2:03:11, 2.84s/it] + Training...: 41% 1796/4393 [2:31:23<1:52:03, 2.59s/it] + Training...: 41% 1797/4393 [2:31:25<1:41:13, 2.34s/it] + Training...: 41% 1798/4393 [2:31:27<1:30:46, 2.10s/it] + Training...: 41% 1799/4393 [2:31:28<1:20:17, 1.86s/it] + Training...: 41% 1800/4393 [2:31:29<1:09:18, 1.60s/it] + Training...: 41% 1801/4393 [2:31:35<2:09:34, 3.00s/it] + Training...: 41% 1802/4393 [2:31:42<2:52:50, 4.00s/it] + Training...: 41% 1803/4393 [2:31:48<3:22:36, 4.69s/it] + Training...: 41% 1804/4393 [2:31:54<3:39:33, 5.09s/it] + Training...: 41% 1805/4393 [2:32:00<3:51:12, 5.36s/it] + Training...: 41% 1806/4393 [2:32:06<3:57:37, 5.51s/it] + Training...: 41% 1807/4393 [2:32:12<4:02:10, 5.62s/it] + Training...: 41% 1808/4393 [2:32:18<4:05:54, 5.71s/it] + Training...: 41% 1809/4393 [2:32:23<4:07:54, 5.76s/it] + Training...: 41% 1810/4393 [2:32:29<4:08:17, 5.77s/it] + Training...: 41% 1811/4393 [2:32:35<4:08:31, 5.78s/it] + Training...: 41% 1812/4393 [2:32:41<4:07:01, 5.74s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:09:47<17:31:17, 21025.84s/it] + Training...: 41% 1812/4393 [2:32:47<4:07:01, 5.74s/it] + Training...: 41% 1813/4393 [2:32:47<4:17:15, 5.98s/it] + Training...: 41% 1814/4393 [2:32:53<4:13:21, 5.89s/it] + Training...: 41% 1815/4393 [2:32:59<4:10:37, 5.83s/it] + Training...: 41% 1816/4393 [2:33:04<4:08:05, 5.78s/it] + Training...: 41% 1817/4393 [2:33:10<4:09:18, 5.81s/it] + Training...: 41% 1818/4393 [2:33:16<4:06:16, 5.74s/it] + Training...: 41% 1819/4393 [2:33:21<4:04:28, 5.70s/it] + Training...: 41% 1820/4393 [2:33:27<4:00:42, 5.61s/it] + Training...: 41% 1821/4393 [2:33:32<3:58:46, 5.57s/it] + Training...: 41% 1822/4393 [2:33:38<3:56:28, 5.52s/it] + Training...: 41% 1823/4393 [2:33:43<3:55:18, 5.49s/it] + Training...: 42% 1824/4393 [2:33:48<3:53:06, 5.44s/it] + Training...: 42% 1825/4393 [2:33:54<3:51:38, 5.41s/it] + Training...: 42% 1826/4393 [2:33:59<3:48:48, 5.35s/it] + Training...: 42% 1827/4393 [2:34:04<3:46:30, 5.30s/it] + Training...: 42% 1828/4393 [2:34:09<3:43:57, 5.24s/it] + Training...: 42% 1829/4393 [2:34:14<3:42:05, 5.20s/it] + Training...: 42% 1830/4393 [2:34:19<3:39:49, 5.15s/it] + Training...: 42% 1831/4393 [2:34:24<3:40:08, 5.16s/it] + Training...: 42% 1832/4393 [2:34:29<3:37:46, 5.10s/it] + Training...: 42% 1833/4393 [2:34:34<3:34:25, 5.03s/it] + Training...: 42% 1834/4393 [2:34:39<3:29:36, 4.91s/it] + Training...: 42% 1835/4393 [2:34:44<3:25:30, 4.82s/it] + Training...: 42% 1836/4393 [2:34:48<3:19:43, 4.69s/it] + Training...: 42% 1837/4393 [2:34:52<3:13:38, 4.55s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:11:57<17:31:17, 21025.84s/it] + Training...: 42% 1837/4393 [2:34:57<3:13:38, 4.55s/it] + Training...: 42% 1838/4393 [2:34:57<3:15:15, 4.59s/it] + Training...: 42% 1839/4393 [2:35:01<3:06:00, 4.37s/it] + Training...: 42% 1840/4393 [2:35:04<2:56:01, 4.14s/it] + Training...: 42% 1841/4393 [2:35:08<2:45:49, 3.90s/it] + Training...: 42% 1842/4393 [2:35:11<2:35:43, 3.66s/it] + Training...: 42% 1843/4393 [2:35:13<2:24:38, 3.40s/it] + Training...: 42% 1844/4393 [2:35:16<2:13:54, 3.15s/it] + Training...: 42% 1845/4393 [2:35:18<2:03:24, 2.91s/it] + Training...: 42% 1846/4393 [2:35:20<1:52:53, 2.66s/it] + Training...: 42% 1847/4393 [2:35:22<1:41:56, 2.40s/it] + Training...: 42% 1848/4393 [2:35:24<1:31:28, 2.16s/it] + Training...: 42% 1849/4393 [2:35:25<1:20:32, 1.90s/it] + Training...: 42% 1850/4393 [2:35:26<1:09:14, 1.63s/it] + Training...: 42% 1851/4393 [2:35:32<2:08:44, 3.04s/it] + Training...: 42% 1852/4393 [2:35:39<2:49:03, 3.99s/it] + Training...: 42% 1853/4393 [2:35:45<3:15:27, 4.62s/it] + Training...: 42% 1854/4393 [2:35:51<3:33:21, 5.04s/it] + Training...: 42% 1855/4393 [2:35:57<3:49:35, 5.43s/it] + Training...: 42% 1856/4393 [2:36:03<3:56:19, 5.59s/it] + Training...: 42% 1857/4393 [2:36:09<4:00:26, 5.69s/it] + Training...: 42% 1858/4393 [2:36:15<4:01:45, 5.72s/it] + Training...: 42% 1859/4393 [2:36:21<4:02:58, 5.75s/it] + Training...: 42% 1860/4393 [2:36:26<4:03:06, 5.76s/it] + Training...: 42% 1861/4393 [2:36:32<4:03:22, 5.77s/it] + Training...: 42% 1862/4393 [2:36:38<4:02:10, 5.74s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:13:44<17:31:17, 21025.84s/it] + Training...: 42% 1862/4393 [2:36:45<4:02:10, 5.74s/it] + Training...: 42% 1863/4393 [2:36:45<4:13:26, 6.01s/it] + Training...: 42% 1864/4393 [2:36:50<4:09:25, 5.92s/it] + Training...: 42% 1865/4393 [2:36:56<4:05:59, 5.84s/it] + Training...: 42% 1866/4393 [2:37:01<4:03:00, 5.77s/it] + Training...: 42% 1867/4393 [2:37:07<4:00:46, 5.72s/it] + Training...: 43% 1868/4393 [2:37:13<3:58:25, 5.67s/it] + Training...: 43% 1869/4393 [2:37:18<3:57:56, 5.66s/it] + Training...: 43% 1870/4393 [2:37:24<3:57:59, 5.66s/it] + Training...: 43% 1871/4393 [2:37:29<3:54:57, 5.59s/it] + Training...: 43% 1872/4393 [2:37:35<3:51:26, 5.51s/it] + Training...: 43% 1873/4393 [2:37:40<3:48:49, 5.45s/it] + Training...: 43% 1874/4393 [2:37:45<3:46:54, 5.40s/it] + Training...: 43% 1875/4393 [2:37:51<3:44:27, 5.35s/it] + Training...: 43% 1876/4393 [2:37:56<3:41:56, 5.29s/it] + Training...: 43% 1877/4393 [2:38:01<3:40:02, 5.25s/it] + Training...: 43% 1878/4393 [2:38:06<3:37:54, 5.20s/it] + Training...: 43% 1879/4393 [2:38:11<3:35:34, 5.15s/it] + Training...: 43% 1880/4393 [2:38:16<3:33:00, 5.09s/it] + Training...: 43% 1881/4393 [2:38:21<3:30:44, 5.03s/it] + Training...: 43% 1882/4393 [2:38:26<3:28:24, 4.98s/it] + Training...: 43% 1883/4393 [2:38:30<3:25:56, 4.92s/it] + Training...: 43% 1884/4393 [2:38:35<3:23:14, 4.86s/it] + Training...: 43% 1885/4393 [2:38:40<3:20:08, 4.79s/it] + Training...: 43% 1886/4393 [2:38:44<3:15:34, 4.68s/it] + Training...: 43% 1887/4393 [2:38:48<3:10:50, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:15:53<17:31:17, 21025.84s/it] + Training...: 43% 1887/4393 [2:38:53<3:10:50, 4.57s/it] + Training...: 43% 1888/4393 [2:38:53<3:13:22, 4.63s/it] + Training...: 43% 1889/4393 [2:38:57<3:05:23, 4.44s/it] + Training...: 43% 1890/4393 [2:39:01<2:55:47, 4.21s/it] + Training...: 43% 1891/4393 [2:39:04<2:45:52, 3.98s/it] + Training...: 43% 1892/4393 [2:39:07<2:34:22, 3.70s/it] + Training...: 43% 1893/4393 [2:39:10<2:24:09, 3.46s/it] + Training...: 43% 1894/4393 [2:39:13<2:13:16, 3.20s/it] + Training...: 43% 1895/4393 [2:39:15<2:02:28, 2.94s/it] + Training...: 43% 1896/4393 [2:39:17<1:52:11, 2.70s/it] + Training...: 43% 1897/4393 [2:39:19<1:41:37, 2.44s/it] + Training...: 43% 1898/4393 [2:39:21<1:30:56, 2.19s/it] + Training...: 43% 1899/4393 [2:39:22<1:20:05, 1.93s/it] + Training...: 43% 1900/4393 [2:39:23<1:08:46, 1.66s/it] + Training...: 43% 1901/4393 [2:39:29<2:04:22, 2.99s/it] + Training...: 43% 1902/4393 [2:39:36<2:44:31, 3.96s/it] + Training...: 43% 1903/4393 [2:39:42<3:10:17, 4.59s/it] + Training...: 43% 1904/4393 [2:39:48<3:27:12, 4.99s/it] + Training...: 43% 1905/4393 [2:39:54<3:39:46, 5.30s/it] + Training...: 43% 1906/4393 [2:39:59<3:47:25, 5.49s/it] + Training...: 43% 1907/4393 [2:40:05<3:52:33, 5.61s/it] + Training...: 43% 1908/4393 [2:40:11<3:55:11, 5.68s/it] + Training...: 43% 1909/4393 [2:40:17<3:58:11, 5.75s/it] + Training...: 43% 1910/4393 [2:40:23<3:59:36, 5.79s/it] + Training...: 44% 1911/4393 [2:40:29<4:03:59, 5.90s/it] + Training...: 44% 1912/4393 [2:40:35<4:01:38, 5.84s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:17:41<17:31:17, 21025.84s/it] + Training...: 44% 1912/4393 [2:40:41<4:01:38, 5.84s/it] + Training...: 44% 1913/4393 [2:40:41<4:11:02, 6.07s/it] + Training...: 44% 1914/4393 [2:40:47<4:06:13, 5.96s/it] + Training...: 44% 1915/4393 [2:40:53<4:02:27, 5.87s/it] + Training...: 44% 1916/4393 [2:40:58<3:58:57, 5.79s/it] + Training...: 44% 1917/4393 [2:41:04<3:57:07, 5.75s/it] + Training...: 44% 1918/4393 [2:41:10<3:54:59, 5.70s/it] + Training...: 44% 1919/4393 [2:41:15<3:53:32, 5.66s/it] + Training...: 44% 1920/4393 [2:41:21<3:51:28, 5.62s/it] + Training...: 44% 1921/4393 [2:41:26<3:48:54, 5.56s/it] + Training...: 44% 1922/4393 [2:41:31<3:45:39, 5.48s/it] + Training...: 44% 1923/4393 [2:41:37<3:44:15, 5.45s/it] + Training...: 44% 1924/4393 [2:41:42<3:42:10, 5.40s/it] + Training...: 44% 1925/4393 [2:41:47<3:40:00, 5.35s/it] + Training...: 44% 1926/4393 [2:41:53<3:39:08, 5.33s/it] + Training...: 44% 1927/4393 [2:41:58<3:39:02, 5.33s/it] + Training...: 44% 1928/4393 [2:42:03<3:35:44, 5.25s/it] + Training...: 44% 1929/4393 [2:42:08<3:33:07, 5.19s/it] + Training...: 44% 1930/4393 [2:42:13<3:30:32, 5.13s/it] + Training...: 44% 1931/4393 [2:42:18<3:27:36, 5.06s/it] + Training...: 44% 1932/4393 [2:42:23<3:24:18, 4.98s/it] + Training...: 44% 1933/4393 [2:42:28<3:22:02, 4.93s/it] + Training...: 44% 1934/4393 [2:42:32<3:18:23, 4.84s/it] + Training...: 44% 1935/4393 [2:42:37<3:15:08, 4.76s/it] + Training...: 44% 1936/4393 [2:42:41<3:11:39, 4.68s/it] + Training...: 44% 1937/4393 [2:42:46<3:06:11, 4.55s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:19:50<17:31:17, 21025.84s/it] + Training...: 44% 1937/4393 [2:42:50<3:06:11, 4.55s/it] + Training...: 44% 1938/4393 [2:42:50<3:07:43, 4.59s/it] + Training...: 44% 1939/4393 [2:42:54<2:59:10, 4.38s/it] + Training...: 44% 1940/4393 [2:42:58<2:50:08, 4.16s/it] + Training...: 44% 1941/4393 [2:43:01<2:41:22, 3.95s/it] + Training...: 44% 1942/4393 [2:43:04<2:32:03, 3.72s/it] + Training...: 44% 1943/4393 [2:43:07<2:22:09, 3.48s/it] + Training...: 44% 1944/4393 [2:43:10<2:11:44, 3.23s/it] + Training...: 44% 1945/4393 [2:43:12<2:01:30, 2.98s/it] + Training...: 44% 1946/4393 [2:43:14<1:50:55, 2.72s/it] + Training...: 44% 1947/4393 [2:43:16<1:40:10, 2.46s/it] + Training...: 44% 1948/4393 [2:43:18<1:28:48, 2.18s/it] + Training...: 44% 1949/4393 [2:43:19<1:17:33, 1.90s/it] + Training...: 44% 1950/4393 [2:43:20<1:06:37, 1.64s/it] + Training...: 44% 1951/4393 [2:43:26<2:02:18, 3.00s/it] + Training...: 44% 1952/4393 [2:43:32<2:40:40, 3.95s/it] + Training...: 44% 1953/4393 [2:43:39<3:06:37, 4.59s/it] + Training...: 44% 1954/4393 [2:43:45<3:23:31, 5.01s/it] + Training...: 45% 1955/4393 [2:43:51<3:35:31, 5.30s/it] + Training...: 45% 1956/4393 [2:43:56<3:43:44, 5.51s/it] + Training...: 45% 1957/4393 [2:44:02<3:49:33, 5.65s/it] + Training...: 45% 1958/4393 [2:44:08<3:53:07, 5.74s/it] + Training...: 45% 1959/4393 [2:44:14<3:56:13, 5.82s/it] + Training...: 45% 1960/4393 [2:44:20<3:56:53, 5.84s/it] + Training...: 45% 1961/4393 [2:44:26<3:55:36, 5.81s/it] + Training...: 45% 1962/4393 [2:44:32<3:54:13, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:21:38<17:31:17, 21025.84s/it] + Training...: 45% 1962/4393 [2:44:38<3:54:13, 5.78s/it] + Training...: 45% 1963/4393 [2:44:38<4:03:15, 6.01s/it] + Training...: 45% 1964/4393 [2:44:44<3:59:32, 5.92s/it] + Training...: 45% 1965/4393 [2:44:50<3:56:13, 5.84s/it] + Training...: 45% 1966/4393 [2:44:55<3:52:46, 5.75s/it] + Training...: 45% 1967/4393 [2:45:01<3:50:02, 5.69s/it] + Training...: 45% 1968/4393 [2:45:06<3:48:00, 5.64s/it] + Training...: 45% 1969/4393 [2:45:12<3:46:26, 5.61s/it] + Training...: 45% 1970/4393 [2:45:17<3:45:18, 5.58s/it] + Training...: 45% 1971/4393 [2:45:23<3:44:35, 5.56s/it] + Training...: 45% 1972/4393 [2:45:28<3:42:08, 5.51s/it] + Training...: 45% 1973/4393 [2:45:34<3:40:06, 5.46s/it] + Training...: 45% 1974/4393 [2:45:39<3:39:04, 5.43s/it] + Training...: 45% 1975/4393 [2:45:44<3:37:36, 5.40s/it] + Training...: 45% 1976/4393 [2:45:49<3:35:13, 5.34s/it] + Training...: 45% 1977/4393 [2:45:55<3:33:00, 5.29s/it] + Training...: 45% 1978/4393 [2:46:00<3:30:36, 5.23s/it] + Training...: 45% 1979/4393 [2:46:05<3:28:36, 5.18s/it] + Training...: 45% 1980/4393 [2:46:10<3:26:34, 5.14s/it] + Training...: 45% 1981/4393 [2:46:15<3:23:46, 5.07s/it] + Training...: 45% 1982/4393 [2:46:20<3:20:58, 5.00s/it] + Training...: 45% 1983/4393 [2:46:24<3:17:59, 4.93s/it] + Training...: 45% 1984/4393 [2:46:29<3:13:56, 4.83s/it] + Training...: 45% 1985/4393 [2:46:34<3:12:39, 4.80s/it] + Training...: 45% 1986/4393 [2:46:38<3:09:52, 4.73s/it] + Training...: 45% 1987/4393 [2:46:43<3:04:47, 4.61s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:23:47<17:31:17, 21025.84s/it] + Training...: 45% 1987/4393 [2:46:47<3:04:47, 4.61s/it] + Training...: 45% 1988/4393 [2:46:47<3:06:06, 4.64s/it] + Training...: 45% 1989/4393 [2:46:51<2:57:09, 4.42s/it] + Training...: 45% 1990/4393 [2:46:55<2:47:12, 4.18s/it] + Training...: 45% 1991/4393 [2:46:58<2:38:10, 3.95s/it] + Training...: 45% 1992/4393 [2:47:01<2:28:57, 3.72s/it] + Training...: 45% 1993/4393 [2:47:04<2:18:43, 3.47s/it] + Training...: 45% 1994/4393 [2:47:07<2:08:34, 3.22s/it] + Training...: 45% 1995/4393 [2:47:09<1:58:13, 2.96s/it] + Training...: 45% 1996/4393 [2:47:11<1:48:24, 2.71s/it] + Training...: 45% 1997/4393 [2:47:13<1:37:58, 2.45s/it] + Training...: 45% 1998/4393 [2:47:15<1:27:26, 2.19s/it] + Training...: 46% 1999/4393 [2:47:16<1:16:17, 1.91s/it] + Training...: 46% 2000/4393 [2:47:17<1:05:03, 1.63s/it] + Training...: 46% 2001/4393 [2:47:23<2:00:22, 3.02s/it] + Training...: 46% 2002/4393 [2:47:30<2:39:32, 4.00s/it] + Training...: 46% 2003/4393 [2:47:36<3:04:05, 4.62s/it] + Training...: 46% 2004/4393 [2:47:42<3:20:29, 5.04s/it] + Training...: 46% 2005/4393 [2:47:48<3:31:41, 5.32s/it] + Training...: 46% 2006/4393 [2:47:54<3:38:04, 5.48s/it] + Training...: 46% 2007/4393 [2:48:00<3:43:50, 5.63s/it] + Training...: 46% 2008/4393 [2:48:05<3:47:31, 5.72s/it] + Training...: 46% 2009/4393 [2:48:11<3:50:36, 5.80s/it] + Training...: 46% 2010/4393 [2:48:17<3:50:13, 5.80s/it] + Training...: 46% 2011/4393 [2:48:23<3:50:06, 5.80s/it] + Training...: 46% 2012/4393 [2:48:29<3:48:47, 5.77s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:25:35<17:31:17, 21025.84s/it] + Training...: 46% 2012/4393 [2:48:35<3:48:47, 5.77s/it] + Training...: 46% 2013/4393 [2:48:35<3:58:13, 6.01s/it] + Training...: 46% 2014/4393 [2:48:41<3:54:40, 5.92s/it] + Training...: 46% 2015/4393 [2:48:47<3:52:29, 5.87s/it] + Training...: 46% 2016/4393 [2:48:52<3:49:23, 5.79s/it] + Training...: 46% 2017/4393 [2:48:58<3:46:32, 5.72s/it] + Training...: 46% 2018/4393 [2:49:03<3:44:04, 5.66s/it] + Training...: 46% 2019/4393 [2:49:09<3:42:06, 5.61s/it] + Training...: 46% 2020/4393 [2:49:15<3:41:57, 5.61s/it] + Training...: 46% 2021/4393 [2:49:20<3:41:40, 5.61s/it] + Training...: 46% 2022/4393 [2:49:26<3:38:29, 5.53s/it] + Training...: 46% 2023/4393 [2:49:31<3:36:39, 5.48s/it] + Training...: 46% 2024/4393 [2:49:36<3:34:35, 5.44s/it] + Training...: 46% 2025/4393 [2:49:41<3:32:27, 5.38s/it] + Training...: 46% 2026/4393 [2:49:47<3:29:43, 5.32s/it] + Training...: 46% 2027/4393 [2:49:52<3:27:41, 5.27s/it] + Training...: 46% 2028/4393 [2:49:57<3:25:21, 5.21s/it] + Training...: 46% 2029/4393 [2:50:02<3:23:03, 5.15s/it] + Training...: 46% 2030/4393 [2:50:07<3:19:53, 5.08s/it] + Training...: 46% 2031/4393 [2:50:12<3:17:45, 5.02s/it] + Training...: 46% 2032/4393 [2:50:16<3:15:00, 4.96s/it] + Training...: 46% 2033/4393 [2:50:21<3:12:44, 4.90s/it] + Training...: 46% 2034/4393 [2:50:26<3:08:19, 4.79s/it] + Training...: 46% 2035/4393 [2:50:30<3:05:19, 4.72s/it] + Training...: 46% 2036/4393 [2:50:35<3:01:18, 4.62s/it] + Training...: 46% 2037/4393 [2:50:39<2:56:45, 4.50s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:27:44<17:31:17, 21025.84s/it] + Training...: 46% 2037/4393 [2:50:44<2:56:45, 4.50s/it] + Training...: 46% 2038/4393 [2:50:44<2:58:43, 4.55s/it] + Training...: 46% 2039/4393 [2:50:47<2:50:16, 4.34s/it] + Training...: 46% 2040/4393 [2:50:51<2:41:28, 4.12s/it] + Training...: 46% 2041/4393 [2:50:54<2:32:23, 3.89s/it] + Training...: 46% 2042/4393 [2:50:58<2:24:48, 3.70s/it] + Training...: 47% 2043/4393 [2:51:01<2:15:08, 3.45s/it] + Training...: 47% 2044/4393 [2:51:03<2:05:23, 3.20s/it] + Training...: 47% 2045/4393 [2:51:06<1:55:28, 2.95s/it] + Training...: 47% 2046/4393 [2:51:08<1:45:55, 2.71s/it] + Training...: 47% 2047/4393 [2:51:10<1:35:43, 2.45s/it] + Training...: 47% 2048/4393 [2:51:11<1:25:37, 2.19s/it] + Training...: 47% 2049/4393 [2:51:12<1:15:20, 1.93s/it] + Training...: 47% 2050/4393 [2:51:13<1:04:53, 1.66s/it] + Training...: 47% 2051/4393 [2:51:20<1:58:39, 3.04s/it] + Training...: 47% 2052/4393 [2:51:26<2:35:45, 3.99s/it] + Training...: 47% 2053/4393 [2:51:32<2:59:15, 4.60s/it] + Training...: 47% 2054/4393 [2:51:38<3:14:55, 5.00s/it] + Training...: 47% 2055/4393 [2:51:44<3:26:41, 5.30s/it] + Training...: 47% 2056/4393 [2:51:50<3:33:40, 5.49s/it] + Training...: 47% 2057/4393 [2:51:56<3:38:08, 5.60s/it] + Training...: 47% 2058/4393 [2:52:02<3:41:16, 5.69s/it] + Training...: 47% 2059/4393 [2:52:07<3:43:44, 5.75s/it] + Training...: 47% 2060/4393 [2:52:13<3:44:13, 5.77s/it] + Training...: 47% 2061/4393 [2:52:19<3:46:44, 5.83s/it] + Training...: 47% 2062/4393 [2:52:25<3:46:39, 5.83s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:29:32<17:31:17, 21025.84s/it] + Training...: 47% 2062/4393 [2:52:32<3:46:39, 5.83s/it] + Training...: 47% 2063/4393 [2:52:32<3:54:16, 6.03s/it] + Training...: 47% 2064/4393 [2:52:37<3:49:49, 5.92s/it] + Training...: 47% 2065/4393 [2:52:43<3:47:02, 5.85s/it] + Training...: 47% 2066/4393 [2:52:49<3:43:34, 5.76s/it] + Training...: 47% 2067/4393 [2:52:54<3:42:10, 5.73s/it] + Training...: 47% 2068/4393 [2:53:00<3:39:12, 5.66s/it] + Training...: 47% 2069/4393 [2:53:05<3:37:50, 5.62s/it] + Training...: 47% 2070/4393 [2:53:11<3:36:13, 5.58s/it] + Training...: 47% 2071/4393 [2:53:16<3:34:25, 5.54s/it] + Training...: 47% 2072/4393 [2:53:22<3:32:35, 5.50s/it] + Training...: 47% 2073/4393 [2:53:27<3:30:58, 5.46s/it] + Training...: 47% 2074/4393 [2:53:32<3:28:06, 5.38s/it] + Training...: 47% 2075/4393 [2:53:37<3:25:46, 5.33s/it] + Training...: 47% 2076/4393 [2:53:42<3:23:34, 5.27s/it] + Training...: 47% 2077/4393 [2:53:47<3:21:08, 5.21s/it] + Training...: 47% 2078/4393 [2:53:53<3:19:13, 5.16s/it] + Training...: 47% 2079/4393 [2:53:58<3:16:49, 5.10s/it] + Training...: 47% 2080/4393 [2:54:02<3:14:59, 5.06s/it] + Training...: 47% 2081/4393 [2:54:07<3:12:47, 5.00s/it] + Training...: 47% 2082/4393 [2:54:12<3:10:51, 4.96s/it] + Training...: 47% 2083/4393 [2:54:17<3:07:46, 4.88s/it] + Training...: 47% 2084/4393 [2:54:21<3:04:13, 4.79s/it] + Training...: 47% 2085/4393 [2:54:26<3:00:45, 4.70s/it] + Training...: 47% 2086/4393 [2:54:30<2:56:23, 4.59s/it] + Training...: 48% 2087/4393 [2:54:34<2:52:05, 4.48s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:31:39<17:31:17, 21025.84s/it] + Training...: 48% 2087/4393 [2:54:39<2:52:05, 4.48s/it] + Training...: 48% 2088/4393 [2:54:39<2:54:47, 4.55s/it] + Training...: 48% 2089/4393 [2:54:43<2:45:36, 4.31s/it] + Training...: 48% 2090/4393 [2:54:46<2:35:58, 4.06s/it] + Training...: 48% 2091/4393 [2:54:50<2:27:06, 3.83s/it] + Training...: 48% 2092/4393 [2:54:53<2:18:00, 3.60s/it] + Training...: 48% 2093/4393 [2:54:56<2:08:56, 3.36s/it] + Training...: 48% 2094/4393 [2:54:58<2:00:06, 3.13s/it] + Training...: 48% 2095/4393 [2:55:01<1:50:57, 2.90s/it] + Training...: 48% 2096/4393 [2:55:03<1:42:03, 2.67s/it] + Training...: 48% 2097/4393 [2:55:05<1:32:33, 2.42s/it] + Training...: 48% 2098/4393 [2:55:06<1:22:56, 2.17s/it] + Training...: 48% 2099/4393 [2:55:07<1:13:06, 1.91s/it] + Training...: 48% 2100/4393 [2:55:08<1:02:36, 1.64s/it] + Training...: 48% 2101/4393 [2:55:15<1:55:11, 3.02s/it] + Training...: 48% 2102/4393 [2:55:21<2:31:45, 3.97s/it] + Training...: 48% 2103/4393 [2:55:27<2:56:06, 4.61s/it] + Training...: 48% 2104/4393 [2:55:33<3:11:28, 5.02s/it] + Training...: 48% 2105/4393 [2:55:39<3:22:17, 5.30s/it] + Training...: 48% 2106/4393 [2:55:45<3:28:41, 5.48s/it] + Training...: 48% 2107/4393 [2:55:51<3:33:55, 5.61s/it] + Training...: 48% 2108/4393 [2:55:57<3:36:33, 5.69s/it] + Training...: 48% 2109/4393 [2:56:03<3:39:26, 5.76s/it] + Training...: 48% 2110/4393 [2:56:08<3:40:34, 5.80s/it] + Training...: 48% 2111/4393 [2:56:14<3:40:50, 5.81s/it] + Training...: 48% 2112/4393 [2:56:20<3:40:07, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:33:26<17:31:17, 21025.84s/it] + Training...: 48% 2112/4393 [2:56:26<3:40:07, 5.79s/it] + Training...: 48% 2113/4393 [2:56:26<3:47:52, 6.00s/it] + Training...: 48% 2114/4393 [2:56:32<3:43:32, 5.89s/it] + Training...: 48% 2115/4393 [2:56:38<3:40:36, 5.81s/it] + Training...: 48% 2116/4393 [2:56:43<3:39:29, 5.78s/it] + Training...: 48% 2117/4393 [2:56:49<3:38:42, 5.77s/it] + Training...: 48% 2118/4393 [2:56:55<3:35:41, 5.69s/it] + Training...: 48% 2119/4393 [2:57:00<3:33:24, 5.63s/it] + Training...: 48% 2120/4393 [2:57:06<3:30:32, 5.56s/it] + Training...: 48% 2121/4393 [2:57:11<3:28:48, 5.51s/it] + Training...: 48% 2122/4393 [2:57:16<3:26:59, 5.47s/it] + Training...: 48% 2123/4393 [2:57:22<3:25:58, 5.44s/it] + Training...: 48% 2124/4393 [2:57:27<3:24:35, 5.41s/it] + Training...: 48% 2125/4393 [2:57:32<3:23:01, 5.37s/it] + Training...: 48% 2126/4393 [2:57:38<3:20:50, 5.32s/it] + Training...: 48% 2127/4393 [2:57:43<3:19:03, 5.27s/it] + Training...: 48% 2128/4393 [2:57:48<3:16:41, 5.21s/it] + Training...: 48% 2129/4393 [2:57:53<3:14:46, 5.16s/it] + Training...: 48% 2130/4393 [2:57:58<3:11:39, 5.08s/it] + Training...: 49% 2131/4393 [2:58:03<3:09:18, 5.02s/it] + Training...: 49% 2132/4393 [2:58:07<3:08:01, 4.99s/it] + Training...: 49% 2133/4393 [2:58:12<3:06:52, 4.96s/it] + Training...: 49% 2134/4393 [2:58:17<3:02:04, 4.84s/it] + Training...: 49% 2135/4393 [2:58:21<2:57:50, 4.73s/it] + Training...: 49% 2136/4393 [2:58:26<2:53:10, 4.60s/it] + Training...: 49% 2137/4393 [2:58:30<2:48:58, 4.49s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:35:35<17:31:17, 21025.84s/it] + Training...: 49% 2137/4393 [2:58:35<2:48:58, 4.49s/it] + Training...: 49% 2138/4393 [2:58:35<2:50:42, 4.54s/it] + Training...: 49% 2139/4393 [2:58:38<2:43:09, 4.34s/it] + Training...: 49% 2140/4393 [2:58:42<2:34:59, 4.13s/it] + Training...: 49% 2141/4393 [2:58:45<2:26:20, 3.90s/it] + Training...: 49% 2142/4393 [2:58:49<2:17:54, 3.68s/it] + Training...: 49% 2143/4393 [2:58:51<2:08:28, 3.43s/it] + Training...: 49% 2144/4393 [2:58:54<1:58:22, 3.16s/it] + Training...: 49% 2145/4393 [2:58:56<1:48:43, 2.90s/it] + Training...: 49% 2146/4393 [2:58:58<1:38:40, 2.63s/it] + Training...: 49% 2147/4393 [2:59:00<1:29:01, 2.38s/it] + Training...: 49% 2148/4393 [2:59:02<1:19:56, 2.14s/it] + Training...: 49% 2149/4393 [2:59:03<1:10:40, 1.89s/it] + Training...: 49% 2150/4393 [2:59:04<1:00:47, 1.63s/it] + Training...: 49% 2151/4393 [2:59:10<1:51:57, 3.00s/it] + Training...: 49% 2152/4393 [2:59:16<2:27:35, 3.95s/it] + Training...: 49% 2153/4393 [2:59:22<2:51:07, 4.58s/it] + Training...: 49% 2154/4393 [2:59:28<3:07:20, 5.02s/it] + Training...: 49% 2155/4393 [2:59:35<3:19:43, 5.35s/it] + Training...: 49% 2156/4393 [2:59:41<3:27:47, 5.57s/it] + Training...: 49% 2157/4393 [2:59:47<3:31:05, 5.66s/it] + Training...: 49% 2158/4393 [2:59:52<3:32:29, 5.70s/it] + Training...: 49% 2159/4393 [2:59:58<3:33:44, 5.74s/it] + Training...: 49% 2160/4393 [3:00:04<3:34:23, 5.76s/it] + Training...: 49% 2161/4393 [3:00:10<3:34:54, 5.78s/it] + Training...: 49% 2162/4393 [3:00:16<3:33:52, 5.75s/it] + Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |) +Step... (40025 | Loss: 0.008939584717154503, Learning Rate: 2.0153533114353195e-05, Gradient Norm: 0.29531925916671753) +Step... (40050 | Loss: 0.013731228187680244, Learning Rate: 2.0103030692553148e-05, Gradient Norm: 0.28097018599510193) +Step... (40075 | Loss: 0.005788063630461693, Learning Rate: 2.00525282707531e-05, Gradient Norm: 0.2137639969587326) +Step... (40100 | Loss: 0.006226960103958845, Learning Rate: 2.000201857299544e-05, Gradient Norm: 0.1285054236650467) +Step... (40125 | Loss: 0.007349521853029728, Learning Rate: 1.9951516151195392e-05, Gradient Norm: 0.233669713139534) +Step... (40150 | Loss: 0.010777842253446579, Learning Rate: 1.9901013729395345e-05, Gradient Norm: 0.24161839485168457) +Step... (40175 | Loss: 0.0042556338012218475, Learning Rate: 1.9850504031637684e-05, Gradient Norm: 0.1810857057571411) +Step... (40200 | Loss: 0.013205939903855324, Learning Rate: 1.9800001609837636e-05, Gradient Norm: 0.2645622491836548) +Step... (40225 | Loss: 0.010763286612927914, Learning Rate: 1.9749491912079975e-05, Gradient Norm: 0.2870539128780365) +Step... (40250 | Loss: 0.012334151193499565, Learning Rate: 1.9698989490279928e-05, Gradient Norm: 0.2389720231294632) +Step... (40275 | Loss: 0.0026933024637401104, Learning Rate: 1.9648485249490477e-05, Gradient Norm: 0.09100257605314255) +Step... (40300 | Loss: 0.007666177581995726, Learning Rate: 1.959797737072222e-05, Gradient Norm: 0.14644892513751984) +Step... (40325 | Loss: 0.006105078849941492, Learning Rate: 1.954747312993277e-05, Gradient Norm: 0.1783716231584549) +Step... (40350 | Loss: 0.007519055623561144, Learning Rate: 1.949697070813272e-05, Gradient Norm: 0.196736142039299) +Step... (40375 | Loss: 0.008998542092740536, Learning Rate: 1.944646101037506e-05, Gradient Norm: 0.26219305396080017) +Step... (40400 | Loss: 0.006001105532050133, Learning Rate: 1.9395958588575013e-05, Gradient Norm: 0.17653429508209229) +Step... (40425 | Loss: 0.003485818626359105, Learning Rate: 1.9345456166774966e-05, Gradient Norm: 0.1488671600818634) +Step... (40450 | Loss: 0.011663154698908329, Learning Rate: 1.9294946469017304e-05, Gradient Norm: 0.21771906316280365) +Step... (40475 | Loss: 0.01133037731051445, Learning Rate: 1.9244444047217257e-05, Gradient Norm: 0.3833330273628235) +Step... (40500 | Loss: 0.004920892883092165, Learning Rate: 1.919394162541721e-05, Gradient Norm: 0.1484016329050064) +Step... (40525 | Loss: 0.016179556027054787, Learning Rate: 1.914343192765955e-05, Gradient Norm: 0.27268052101135254) +Step... (40550 | Loss: 0.007338401861488819, Learning Rate: 1.90929295058595e-05, Gradient Norm: 0.18785510957241058) +Step... (40575 | Loss: 0.003818551544100046, Learning Rate: 1.904242526507005e-05, Gradient Norm: 0.1895657181739807) +Step... (40600 | Loss: 0.008862589485943317, Learning Rate: 1.8991917386301793e-05, Gradient Norm: 0.2292691171169281) +Step... (40625 | Loss: 0.0069699776358902454, Learning Rate: 1.8941413145512342e-05, Gradient Norm: 0.1485799252986908) +Step... (40650 | Loss: 0.010429407469928265, Learning Rate: 1.8890910723712295e-05, Gradient Norm: 0.23329854011535645) +Step... (40675 | Loss: 0.005073365289717913, Learning Rate: 1.8840401025954634e-05, Gradient Norm: 0.1328592151403427) +Step... (40700 | Loss: 0.0052896239794790745, Learning Rate: 1.8789898604154587e-05, Gradient Norm: 0.1811048984527588) +Step... (40725 | Loss: 0.0049026296474039555, Learning Rate: 1.873939618235454e-05, Gradient Norm: 0.1749919205904007) +Step... (40750 | Loss: 0.004890045616775751, Learning Rate: 1.8688886484596878e-05, Gradient Norm: 0.15074510872364044) +Step... (40775 | Loss: 0.003988181706517935, Learning Rate: 1.863838406279683e-05, Gradient Norm: 0.19411082565784454) +Step... (40800 | Loss: 0.00978496391326189, Learning Rate: 1.8587881640996784e-05, Gradient Norm: 0.20905667543411255) +Step... (40825 | Loss: 0.006088871043175459, Learning Rate: 1.8537371943239123e-05, Gradient Norm: 0.3005000054836273) +Step... (4085 Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:37:22<17:31:17, 21025.84s/it] + Training...: 49% 2162/4393 [3:00:22<3:33:52, 5.75s/it] + Training...: 49% 2163/4393 [3:00:22<3:42:39, 5.99s/it]0 | Loss: 0.010514942929148674, Learning Rate: 1.8486869521439075e-05, Gradient Norm: 0.2221343070268631) +Step... (40875 | Loss: 0.006962936837226152, Learning Rate: 1.8436365280649625e-05, Gradient Norm: 0.19761787354946136) +Step... (40900 | Loss: 0.011204604059457779, Learning Rate: 1.8385857401881367e-05, Gradient Norm: 0.25883570313453674) +Step... (40925 | Loss: 0.005226748064160347, Learning Rate: 1.8335353161091916e-05, Gradient Norm: 0.2500029504299164) +Step... (40950 | Loss: 0.010302098467946053, Learning Rate: 1.828485073929187e-05, Gradient Norm: 0.18769465386867523) +Step... (40975 | Loss: 0.005125986412167549, Learning Rate: 1.8234341041534208e-05, Gradient Norm: 0.2729404866695404) +Step... (41000 | Loss: 0.010441966354846954, Learning Rate: 1.818383861973416e-05, Gradient Norm: 0.3461083769798279) +Step... (41025 | Loss: 0.0025685643777251244, Learning Rate: 1.8133336197934113e-05, Gradient Norm: 0.12441041320562363) +Step... (41050 | Loss: 0.005768472328782082, Learning Rate: 1.8082826500176452e-05, Gradient Norm: 0.16698002815246582) +Step... (41075 | Loss: 0.00754266232252121, Learning Rate: 1.8032324078376405e-05, Gradient Norm: 0.2963465750217438) +Step... (41100 | Loss: 0.010014562867581844, Learning Rate: 1.7981821656576358e-05, Gradient Norm: 0.16458502411842346) +Step... (41125 | Loss: 0.0019032791024073958, Learning Rate: 1.7931311958818696e-05, Gradient Norm: 0.06771261245012283) +Step... (41150 | Loss: 0.007408216595649719, Learning Rate: 1.788080953701865e-05, Gradient Norm: 0.1833212673664093) +Step... (41175 | Loss: 0.006798009388148785, Learning Rate: 1.78303052962292e-05, Gradient Norm: 0.28412139415740967) +Step... (41200 | Loss: 0.010547134093940258, Learning Rate: 1.777979741746094e-05, Gradient Norm: 0.6392953991889954) +Step... (41225 | Loss: 0.002890239469707012, Learning Rate: 1.772929317667149e-05, Gradient Norm: 0.14131464064121246) +Step... (41250 | Loss: 0.012237129732966423, Learning Rate: 1.7678790754871443e-05, Gradient Norm: 0.24268372356891632) +Step... (41275 | Loss: 0.003835662268102169, Learning Rate: 1.762828105711378e-05, Gradient Norm: 0.20543915033340454) +Step... (41300 | Loss: 0.009825979359447956, Learning Rate: 1.7577778635313734e-05, Gradient Norm: 0.2134653478860855) +Step... (41325 | Loss: 0.0023147186730057, Learning Rate: 1.7527270756545477e-05, Gradient Norm: 0.10707230120897293) +Step... (41350 | Loss: 0.0070909978821873665, Learning Rate: 1.7476766515756026e-05, Gradient Norm: 0.17582404613494873) +Step... (41375 | Loss: 0.0033097451087087393, Learning Rate: 1.742626409395598e-05, Gradient Norm: 0.13646838068962097) +Step... (41400 | Loss: 0.008395623415708542, Learning Rate: 1.7375754396198317e-05, Gradient Norm: 0.1624133586883545) +Step... (41425 | Loss: 0.004059635568410158, Learning Rate: 1.732525197439827e-05, Gradient Norm: 0.21894600987434387) +Step... (41450 | Loss: 0.00695464015007019, Learning Rate: 1.7274749552598223e-05, Gradient Norm: 0.1792638897895813) +Step... (41475 | Loss: 0.009820655919611454, Learning Rate: 1.7224239854840562e-05, Gradient Norm: 0.4551573097705841) +Step... (41500 | Loss: 0.006216324865818024, Learning Rate: 1.7173737433040515e-05, Gradient Norm: 0.17348088324069977) +Step... (41525 | Loss: 0.008928379975259304, Learning Rate: 1.7123233192251064e-05, Gradient Norm: 0.32115671038627625) +Step... (41550 | Loss: 0.012027319520711899, Learning Rate: 1.7072725313482806e-05, Gradient Norm: 0.2573080360889435) +Step... (41575 | Loss: 0.007203887682408094, Learning Rate: 1.7022221072693355e-05, Gradient Norm: 0.24148206412792206) +Step... (41600 | Loss: 0.01733251102268696, Learning Rate: 1.6971718650893308e-05, Gradient Norm: 0.28181859850883484) +Step... (41625 | Loss: 0.00894920527935028, Learning Rate: 1.6921208953135647e-05, Gradient Norm: 0.282490611076355) +Step... (41650 | Loss: 0.009486709721386433, Learning Rate: 1.68707065313356e-05, Gradient Norm: 0.2058715522289276) +Step... (41675 | Loss: 0.020942404866218567, Learning Rate: 1.6820204109535553e-05, Gradient Norm: 0.899817705154419) + + Training...: 49% 2164/4393 [3:00:28<3:42:12, 5.98s/it] + Training...: 49% 2165/4393 [3:00:34<3:38:21, 5.88s/it] + Training...: 49% 2166/4393 [3:00:39<3:34:04, 5.77s/it] + Training...: 49% 2167/4393 [3:00:45<3:31:05, 5.69s/it] + Training...: 49% 2168/4393 [3:00:50<3:28:52, 5.63s/it] + Training...: 49% 2169/4393 [3:00:56<3:26:53, 5.58s/it] + Training...: 49% 2170/4393 [3:01:01<3:26:22, 5.57s/it] + Training...: 49% 2171/4393 [3:01:07<3:26:21, 5.57s/it] + Training...: 49% 2172/4393 [3:01:12<3:24:19, 5.52s/it] + Training...: 49% 2173/4393 [3:01:17<3:22:09, 5.46s/it] + Training...: 49% 2174/4393 [3:01:23<3:20:05, 5.41s/it] + Training...: 50% 2175/4393 [3:01:28<3:18:50, 5.38s/it] + Training...: 50% 2176/4393 [3:01:33<3:16:36, 5.32s/it] + Training...: 50% 2177/4393 [3:01:38<3:14:33, 5.27s/it] + Training...: 50% 2178/4393 [3:01:43<3:12:17, 5.21s/it] + Training...: 50% 2179/4393 [3:01:49<3:10:45, 5.17s/it] + Training...: 50% 2180/4393 [3:01:54<3:08:33, 5.11s/it] + Training...: 50% 2181/4393 [3:01:58<3:06:47, 5.07s/it] + Training...: 50% 2182/4393 [3:02:03<3:03:48, 4.99s/it] + Training...: 50% 2183/4393 [3:02:08<3:01:38, 4.93s/it] + Training...: 50% 2184/4393 [3:02:13<2:58:19, 4.84s/it] + Training...: 50% 2185/4393 [3:02:17<2:55:31, 4.77s/it] + Training...: 50% 2186/4393 [3:02:22<2:51:41, 4.67s/it] + Training...: 50% 2187/4393 [3:02:26<2:47:54, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:39:31<17:31:17, 21025.84s/it] + Training...: 50% 2187/4393 [3:02:31<2:47:54, 4.57s/it] + Training...: 50% 2188/4393 [3:02:31<2:50:25, 4.64s/it] + Training...: 50% 2189/4393 [3:02:35<2:43:18, 4.45s/it] + Training...: 50% 2190/4393 [3:02:39<2:35:14, 4.23s/it] + Training...: 50% 2191/4393 [3:02:42<2:28:41, 4.05s/it] + Training...: 50% 2192/4393 [3:02:46<2:20:19, 3.83s/it] + Training...: 50% 2193/4393 [3:02:49<2:11:20, 3.58s/it] + Training...: 50% 2194/4393 [3:02:51<2:01:41, 3.32s/it] + Training...: 50% 2195/4393 [3:02:54<1:51:15, 3.04s/it] + Training...: 50% 2196/4393 [3:02:56<1:41:27, 2.77s/it] + Training...: 50% 2197/4393 [3:02:58<1:31:07, 2.49s/it] + Training...: 50% 2198/4393 [3:02:59<1:20:40, 2.21s/it] + Training...: 50% 2199/4393 [3:03:00<1:10:18, 1.92s/it] + Training...: 50% 2200/4393 [3:03:01<1:00:15, 1.65s/it] + Training...: 50% 2201/4393 [3:03:08<1:50:49, 3.03s/it] + Training...: 50% 2202/4393 [3:03:14<2:25:46, 3.99s/it] + Training...: 50% 2203/4393 [3:03:20<2:49:06, 4.63s/it] + Training...: 50% 2204/4393 [3:03:26<3:04:58, 5.07s/it] + Training...: 50% 2205/4393 [3:03:32<3:14:50, 5.34s/it] + Training...: 50% 2206/4393 [3:03:38<3:20:15, 5.49s/it] + Training...: 50% 2207/4393 [3:03:44<3:24:33, 5.61s/it] + Training...: 50% 2208/4393 [3:03:50<3:27:13, 5.69s/it] + Training...: 50% 2209/4393 [3:03:56<3:28:54, 5.74s/it] + Training...: 50% 2210/4393 [3:04:01<3:29:40, 5.76s/it] + Training...: 50% 2211/4393 [3:04:07<3:30:23, 5.79s/it] + Training...: 50% 2212/4393 [3:04:13<3:30:00, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:41:19<17:31:17, 21025.84s/it] + Training...: 50% 2212/4393 [3:04:20<3:30:00, 5.78s/it] + Training...: 50% 2213/4393 [3:04:20<3:37:50, 6.00s/it] + Training...: 50% 2214/4393 [3:04:25<3:34:12, 5.90s/it] + Training...: 50% 2215/4393 [3:04:31<3:31:57, 5.84s/it] + Training...: 50% 2216/4393 [3:04:37<3:29:29, 5.77s/it] + Training...: 50% 2217/4393 [3:04:42<3:27:16, 5.72s/it] + Training...: 50% 2218/4393 [3:04:48<3:24:48, 5.65s/it] + Training...: 51% 2219/4393 [3:04:53<3:23:17, 5.61s/it] + Training...: 51% 2220/4393 [3:04:59<3:21:42, 5.57s/it] + Training...: 51% 2221/4393 [3:05:04<3:21:05, 5.55s/it] + Training...: 51% 2222/4393 [3:05:10<3:20:06, 5.53s/it] + Training...: 51% 2223/4393 [3:05:15<3:18:19, 5.48s/it] + Training...: 51% 2224/4393 [3:05:20<3:16:47, 5.44s/it] + Training...: 51% 2225/4393 [3:05:26<3:15:22, 5.41s/it] + Training...: 51% 2226/4393 [3:05:31<3:13:05, 5.35s/it] + Training...: 51% 2227/4393 [3:05:36<3:10:55, 5.29s/it] + Training...: 51% 2228/4393 [3:05:41<3:09:06, 5.24s/it] + Training...: 51% 2229/4393 [3:05:46<3:06:56, 5.18s/it] + Training...: 51% 2230/4393 [3:05:51<3:04:18, 5.11s/it] + Training...: 51% 2231/4393 [3:05:56<3:02:05, 5.05s/it] + Training...: 51% 2232/4393 [3:06:01<2:59:43, 4.99s/it] + Training...: 51% 2233/4393 [3:06:06<2:57:29, 4.93s/it] + Training...: 51% 2234/4393 [3:06:10<2:54:45, 4.86s/it] + Training...: 51% 2235/4393 [3:06:15<2:51:27, 4.77s/it] + Training...: 51% 2236/4393 [3:06:19<2:48:08, 4.68s/it] + Training...: 51% 2237/4393 [3:06:24<2:43:46, 4.56s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:43:28<17:31:17, 21025.84s/it] + Training...: 51% 2237/4393 [3:06:28<2:43:46, 4.56s/it] + Training...: 51% 2238/4393 [3:06:28<2:45:29, 4.61s/it] + Training...: 51% 2239/4393 [3:06:32<2:38:22, 4.41s/it] + Training...: 51% 2240/4393 [3:06:36<2:30:40, 4.20s/it] + Training...: 51% 2241/4393 [3:06:39<2:22:16, 3.97s/it] + Training...: 51% 2242/4393 [3:06:43<2:14:09, 3.74s/it] + Training...: 51% 2243/4393 [3:06:46<2:06:01, 3.52s/it] + Training...: 51% 2244/4393 [3:06:48<1:56:57, 3.27s/it] + Training...: 51% 2245/4393 [3:06:51<1:47:10, 2.99s/it] + Training...: 51% 2246/4393 [3:06:53<1:37:07, 2.71s/it] + Training...: 51% 2247/4393 [3:06:55<1:27:24, 2.44s/it] + Training...: 51% 2248/4393 [3:06:56<1:18:25, 2.19s/it] + Training...: 51% 2249/4393 [3:06:58<1:09:06, 1.93s/it] + Training...: 51% 2250/4393 [3:06:59<59:04, 1.65s/it]  + Training...: 51% 2251/4393 [3:07:05<1:50:15, 3.09s/it] + Training...: 51% 2252/4393 [3:07:11<2:24:10, 4.04s/it] + Training...: 51% 2253/4393 [3:07:17<2:45:51, 4.65s/it] + Training...: 51% 2254/4393 [3:07:23<2:59:56, 5.05s/it] + Training...: 51% 2255/4393 [3:07:29<3:09:32, 5.32s/it] + Training...: 51% 2256/4393 [3:07:35<3:15:28, 5.49s/it] + Training...: 51% 2257/4393 [3:07:41<3:19:48, 5.61s/it] + Training...: 51% 2258/4393 [3:07:47<3:22:31, 5.69s/it] + Training...: 51% 2259/4393 [3:07:53<3:24:50, 5.76s/it] + Training...: 51% 2260/4393 [3:07:59<3:24:59, 5.77s/it] + Training...: 51% 2261/4393 [3:08:04<3:24:45, 5.76s/it] + Training...: 51% 2262/4393 [3:08:10<3:24:43, 5.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:45:17<17:31:17, 21025.84s/it] + Training...: 51% 2262/4393 [3:08:17<3:24:43, 5.76s/it] + Training...: 52% 2263/4393 [3:08:17<3:32:36, 5.99s/it] + Training...: 52% 2264/4393 [3:08:22<3:29:17, 5.90s/it] + Training...: 52% 2265/4393 [3:08:28<3:26:28, 5.82s/it] + Training...: 52% 2266/4393 [3:08:34<3:23:40, 5.75s/it] + Training...: 52% 2267/4393 [3:08:39<3:21:19, 5.68s/it] + Training...: 52% 2268/4393 [3:08:45<3:19:16, 5.63s/it] + Training...: 52% 2269/4393 [3:08:50<3:18:46, 5.62s/it] + Training...: 52% 2270/4393 [3:08:56<3:17:49, 5.59s/it] + Training...: 52% 2271/4393 [3:09:01<3:16:09, 5.55s/it] + Training...: 52% 2272/4393 [3:09:06<3:14:19, 5.50s/it] + Training...: 52% 2273/4393 [3:09:12<3:12:37, 5.45s/it] + Training...: 52% 2274/4393 [3:09:17<3:10:30, 5.39s/it] + Training...: 52% 2275/4393 [3:09:22<3:08:47, 5.35s/it] + Training...: 52% 2276/4393 [3:09:28<3:07:46, 5.32s/it] + Training...: 52% 2277/4393 [3:09:33<3:06:23, 5.29s/it] + Training...: 52% 2278/4393 [3:09:38<3:04:46, 5.24s/it] + Training...: 52% 2279/4393 [3:09:43<3:03:14, 5.20s/it] + Training...: 52% 2280/4393 [3:09:48<3:00:22, 5.12s/it] + Training...: 52% 2281/4393 [3:09:53<2:58:38, 5.08s/it] + Training...: 52% 2282/4393 [3:09:58<2:56:01, 5.00s/it] + Training...: 52% 2283/4393 [3:10:03<2:53:17, 4.93s/it] + Training...: 52% 2284/4393 [3:10:07<2:49:43, 4.83s/it] + Training...: 52% 2285/4393 [3:10:12<2:46:37, 4.74s/it] + Training...: 52% 2286/4393 [3:10:16<2:41:57, 4.61s/it] + Training...: 52% 2287/4393 [3:10:20<2:37:39, 4.49s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:47:25<17:31:17, 21025.84s/it] + Training...: 52% 2287/4393 [3:10:25<2:37:39, 4.49s/it] + Training...: 52% 2288/4393 [3:10:25<2:39:17, 4.54s/it] + Training...: 52% 2289/4393 [3:10:29<2:31:42, 4.33s/it] + Training...: 52% 2290/4393 [3:10:32<2:23:42, 4.10s/it] + Training...: 52% 2291/4393 [3:10:36<2:15:56, 3.88s/it] + Training...: 52% 2292/4393 [3:10:39<2:07:40, 3.65s/it] + Training...: 52% 2293/4393 [3:10:42<1:59:47, 3.42s/it] + Training...: 52% 2294/4393 [3:10:44<1:51:44, 3.19s/it] + Training...: 52% 2295/4393 [3:10:47<1:42:12, 2.92s/it] + Training...: 52% 2296/4393 [3:10:49<1:33:11, 2.67s/it] + Training...: 52% 2297/4393 [3:10:50<1:24:00, 2.40s/it] + Training...: 52% 2298/4393 [3:10:52<1:14:54, 2.15s/it] + Training...: 52% 2299/4393 [3:10:53<1:05:27, 1.88s/it] + Training...: 52% 2300/4393 [3:10:54<56:24, 1.62s/it]  + Training...: 52% 2301/4393 [3:11:00<1:44:10, 2.99s/it] + Training...: 52% 2302/4393 [3:11:07<2:18:26, 3.97s/it] + Training...: 52% 2303/4393 [3:11:13<2:40:41, 4.61s/it] + Training...: 52% 2304/4393 [3:11:19<2:55:11, 5.03s/it] + Training...: 52% 2305/4393 [3:11:25<3:05:16, 5.32s/it] + Training...: 52% 2306/4393 [3:11:31<3:11:31, 5.51s/it] + Training...: 53% 2307/4393 [3:11:37<3:16:08, 5.64s/it] + Training...: 53% 2308/4393 [3:11:43<3:18:30, 5.71s/it] + Training...: 53% 2309/4393 [3:11:48<3:20:01, 5.76s/it] + Training...: 53% 2310/4393 [3:11:54<3:20:04, 5.76s/it] + Training...: 53% 2311/4393 [3:12:00<3:20:20, 5.77s/it] + Training...: 53% 2312/4393 [3:12:06<3:19:54, 5.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:49:12<17:31:17, 21025.84s/it] + Training...: 53% 2312/4393 [3:12:12<3:19:54, 5.76s/it] + Training...: 53% 2313/4393 [3:12:12<3:28:37, 6.02s/it] + Training...: 53% 2314/4393 [3:12:18<3:26:29, 5.96s/it] + Training...: 53% 2315/4393 [3:12:24<3:26:08, 5.95s/it] + Training...: 53% 2316/4393 [3:12:30<3:22:30, 5.85s/it] + Training...: 53% 2317/4393 [3:12:35<3:19:29, 5.77s/it] + Training...: 53% 2318/4393 [3:12:41<3:16:59, 5.70s/it] + Training...: 53% 2319/4393 [3:12:46<3:15:23, 5.65s/it] + Training...: 53% 2320/4393 [3:12:52<3:13:05, 5.59s/it] + Training...: 53% 2321/4393 [3:12:57<3:11:23, 5.54s/it] + Training...: 53% 2322/4393 [3:13:03<3:09:10, 5.48s/it] + Training...: 53% 2323/4393 [3:13:08<3:07:59, 5.45s/it] + Training...: 53% 2324/4393 [3:13:13<3:05:40, 5.38s/it] + Training...: 53% 2325/4393 [3:13:19<3:04:44, 5.36s/it] + Training...: 53% 2326/4393 [3:13:24<3:03:27, 5.33s/it] + Training...: 53% 2327/4393 [3:13:29<3:02:28, 5.30s/it] + Training...: 53% 2328/4393 [3:13:34<3:00:44, 5.25s/it] + Training...: 53% 2329/4393 [3:13:39<2:59:05, 5.21s/it] + Training...: 53% 2330/4393 [3:13:44<2:57:16, 5.16s/it] + Training...: 53% 2331/4393 [3:13:49<2:54:56, 5.09s/it] + Training...: 53% 2332/4393 [3:13:54<2:52:40, 5.03s/it] + Training...: 53% 2333/4393 [3:13:59<2:51:32, 5.00s/it] + Training...: 53% 2334/4393 [3:14:04<2:49:15, 4.93s/it] + Training...: 53% 2335/4393 [3:14:08<2:46:17, 4.85s/it] + Training...: 53% 2336/4393 [3:14:13<2:42:46, 4.75s/it] + Training...: 53% 2337/4393 [3:14:17<2:37:51, 4.61s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:51:22<17:31:17, 21025.84s/it] + Training...: 53% 2337/4393 [3:14:22<2:37:51, 4.61s/it] + Training...: 53% 2338/4393 [3:14:22<2:38:49, 4.64s/it] + Training...: 53% 2339/4393 [3:14:26<2:31:02, 4.41s/it] + Training...: 53% 2340/4393 [3:14:29<2:22:48, 4.17s/it] + Training...: 53% 2341/4393 [3:14:33<2:14:32, 3.93s/it] + Training...: 53% 2342/4393 [3:14:36<2:06:03, 3.69s/it] + Training...: 53% 2343/4393 [3:14:39<1:57:21, 3.43s/it] + Training...: 53% 2344/4393 [3:14:41<1:48:35, 3.18s/it] + Training...: 53% 2345/4393 [3:14:44<1:38:49, 2.90s/it] + Training...: 53% 2346/4393 [3:14:46<1:29:53, 2.64s/it] + Training...: 53% 2347/4393 [3:14:47<1:21:31, 2.39s/it] + Training...: 53% 2348/4393 [3:14:49<1:13:06, 2.15s/it] + Training...: 53% 2349/4393 [3:14:50<1:04:22, 1.89s/it] + Training...: 53% 2350/4393 [3:14:51<55:24, 1.63s/it]  + Training...: 54% 2351/4393 [3:14:58<1:42:08, 3.00s/it] + Training...: 54% 2352/4393 [3:15:04<2:14:50, 3.96s/it] + Training...: 54% 2353/4393 [3:15:10<2:37:04, 4.62s/it] + Training...: 54% 2354/4393 [3:15:16<2:50:32, 5.02s/it] + Training...: 54% 2355/4393 [3:15:22<3:00:02, 5.30s/it] + Training...: 54% 2356/4393 [3:15:28<3:05:55, 5.48s/it] + Training...: 54% 2357/4393 [3:15:34<3:10:32, 5.62s/it] + Training...: 54% 2358/4393 [3:15:39<3:12:47, 5.68s/it] + Training...: 54% 2359/4393 [3:15:45<3:14:45, 5.75s/it] + Training...: 54% 2360/4393 [3:15:51<3:15:11, 5.76s/it] + Training...: 54% 2361/4393 [3:15:57<3:15:33, 5.77s/it] + Training...: 54% 2362/4393 [3:16:03<3:15:48, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:53:09<17:31:17, 21025.84s/it] + Training...: 54% 2362/4393 [3:16:09<3:15:48, 5.78s/it] + Training...: 54% 2363/4393 [3:16:09<3:24:54, 6.06s/it] + Training...: 54% 2364/4393 [3:16:15<3:21:44, 5.97s/it] + Training...: 54% 2365/4393 [3:16:21<3:18:29, 5.87s/it] + Training...: 54% 2366/4393 [3:16:26<3:14:47, 5.77s/it] + Training...: 54% 2367/4393 [3:16:32<3:13:41, 5.74s/it] + Training...: 54% 2368/4393 [3:16:38<3:10:49, 5.65s/it] + Training...: 54% 2369/4393 [3:16:43<3:09:39, 5.62s/it] + Training...: 54% 2370/4393 [3:16:48<3:07:01, 5.55s/it] + Training...: 54% 2371/4393 [3:16:54<3:05:59, 5.52s/it] + Training...: 54% 2372/4393 [3:16:59<3:03:48, 5.46s/it] + Training...: 54% 2373/4393 [3:17:05<3:02:15, 5.41s/it] + Training...: 54% 2374/4393 [3:17:10<3:00:43, 5.37s/it] + Training...: 54% 2375/4393 [3:17:15<3:00:04, 5.35s/it] + Training...: 54% 2376/4393 [3:17:20<2:57:56, 5.29s/it] + Training...: 54% 2377/4393 [3:17:25<2:56:06, 5.24s/it] + Training...: 54% 2378/4393 [3:17:30<2:54:30, 5.20s/it] + Training...: 54% 2379/4393 [3:17:36<2:53:08, 5.16s/it] + Training...: 54% 2380/4393 [3:17:40<2:50:56, 5.10s/it] + Training...: 54% 2381/4393 [3:17:45<2:48:39, 5.03s/it] + Training...: 54% 2382/4393 [3:17:50<2:45:47, 4.95s/it] + Training...: 54% 2383/4393 [3:17:55<2:43:00, 4.87s/it] + Training...: 54% 2384/4393 [3:17:59<2:40:02, 4.78s/it] + Training...: 54% 2385/4393 [3:18:04<2:37:20, 4.70s/it] + Training...: 54% 2386/4393 [3:18:08<2:32:46, 4.57s/it] + Training...: 54% 2387/4393 [3:18:12<2:28:44, 4.45s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:55:17<17:31:17, 21025.84s/it] + Training...: 54% 2387/4393 [3:18:17<2:28:44, 4.45s/it] + Training...: 54% 2388/4393 [3:18:17<2:29:44, 4.48s/it] + Training...: 54% 2389/4393 [3:18:21<2:22:16, 4.26s/it] + Training...: 54% 2390/4393 [3:18:24<2:14:21, 4.02s/it] + Training...: 54% 2391/4393 [3:18:27<2:06:36, 3.79s/it] + Training...: 54% 2392/4393 [3:18:30<1:58:39, 3.56s/it] + Training...: 54% 2393/4393 [3:18:33<1:50:52, 3.33s/it] + Training...: 54% 2394/4393 [3:18:36<1:42:58, 3.09s/it] + Training...: 55% 2395/4393 [3:18:38<1:35:13, 2.86s/it] + Training...: 55% 2396/4393 [3:18:40<1:27:27, 2.63s/it] + Training...: 55% 2397/4393 [3:18:42<1:19:46, 2.40s/it] + Training...: 55% 2398/4393 [3:18:44<1:11:31, 2.15s/it] + Training...: 55% 2399/4393 [3:18:45<1:03:22, 1.91s/it] + Training...: 55% 2400/4393 [3:18:46<54:27, 1.64s/it]  + Training...: 55% 2401/4393 [3:18:52<1:42:38, 3.09s/it] + Training...: 55% 2402/4393 [3:18:59<2:14:22, 4.05s/it] + Training...: 55% 2403/4393 [3:19:05<2:34:55, 4.67s/it] + Training...: 55% 2404/4393 [3:19:11<2:47:59, 5.07s/it] + Training...: 55% 2405/4393 [3:19:17<2:57:29, 5.36s/it] + Training...: 55% 2406/4393 [3:19:23<3:03:03, 5.53s/it] + Training...: 55% 2407/4393 [3:19:29<3:06:48, 5.64s/it] + Training...: 55% 2408/4393 [3:19:35<3:09:02, 5.71s/it] + Training...: 55% 2409/4393 [3:19:40<3:10:20, 5.76s/it] + Training...: 55% 2410/4393 [3:19:46<3:11:05, 5.78s/it] + Training...: 55% 2411/4393 [3:19:52<3:11:13, 5.79s/it] + Training...: 55% 2412/4393 [3:19:58<3:10:39, 5.77s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:57:04<17:31:17, 21025.84s/it] + Training...: 55% 2412/4393 [3:20:04<3:10:39, 5.77s/it] + Training...: 55% 2413/4393 [3:20:04<3:17:59, 6.00s/it] + Training...: 55% 2414/4393 [3:20:10<3:15:28, 5.93s/it] + Training...: 55% 2415/4393 [3:20:16<3:12:45, 5.85s/it] + Training...: 55% 2416/4393 [3:20:21<3:10:15, 5.77s/it] + Training...: 55% 2417/4393 [3:20:27<3:07:52, 5.70s/it] + Training...: 55% 2418/4393 [3:20:32<3:05:53, 5.65s/it] + Training...: 55% 2419/4393 [3:20:38<3:06:09, 5.66s/it] + Training...: 55% 2420/4393 [3:20:44<3:05:39, 5.65s/it] + Training...: 55% 2421/4393 [3:20:49<3:03:33, 5.59s/it] + Training...: 55% 2422/4393 [3:20:54<3:01:24, 5.52s/it] + Training...: 55% 2423/4393 [3:21:00<2:59:36, 5.47s/it] + Training...: 55% 2424/4393 [3:21:05<2:57:52, 5.42s/it] + Training...: 55% 2425/4393 [3:21:10<2:56:45, 5.39s/it] + Training...: 55% 2426/4393 [3:21:16<2:54:38, 5.33s/it] + Training...: 55% 2427/4393 [3:21:21<2:53:09, 5.28s/it] + Training...: 55% 2428/4393 [3:21:26<2:51:06, 5.22s/it] + Training...: 55% 2429/4393 [3:21:31<2:49:43, 5.19s/it] + Training...: 55% 2430/4393 [3:21:36<2:48:02, 5.14s/it] + Training...: 55% 2431/4393 [3:21:41<2:46:14, 5.08s/it] + Training...: 55% 2432/4393 [3:21:46<2:43:40, 5.01s/it] + Training...: 55% 2433/4393 [3:21:51<2:41:22, 4.94s/it] + Training...: 55% 2434/4393 [3:21:55<2:38:26, 4.85s/it] + Training...: 55% 2435/4393 [3:22:00<2:35:42, 4.77s/it] + Training...: 55% 2436/4393 [3:22:04<2:32:03, 4.66s/it] + Training...: 55% 2437/4393 [3:22:08<2:28:05, 4.54s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [55:59:13<17:31:17, 21025.84s/it] + Training...: 55% 2437/4393 [3:22:13<2:28:05, 4.54s/it] + Training...: 55% 2438/4393 [3:22:13<2:29:21, 4.58s/it] + Training...: 56% 2439/4393 [3:22:17<2:22:28, 4.37s/it] + Training...: 56% 2440/4393 [3:22:21<2:14:46, 4.14s/it] + Training...: 56% 2441/4393 [3:22:24<2:07:12, 3.91s/it] + Training...: 56% 2442/4393 [3:22:27<1:58:43, 3.65s/it] + Training...: 56% 2443/4393 [3:22:30<1:50:34, 3.40s/it] + Training...: 56% 2444/4393 [3:22:32<1:42:40, 3.16s/it] + Training...: 56% 2445/4393 [3:22:35<1:33:33, 2.88s/it] + Training...: 56% 2446/4393 [3:22:37<1:24:55, 2.62s/it] + Training...: 56% 2447/4393 [3:22:38<1:16:18, 2.35s/it] + Training...: 56% 2448/4393 [3:22:40<1:08:02, 2.10s/it] + Training...: 56% 2449/4393 [3:22:41<1:00:08, 1.86s/it] + Training...: 56% 2450/4393 [3:22:42<51:42, 1.60s/it]  + Training...: 56% 2451/4393 [3:22:49<1:37:31, 3.01s/it] + Training...: 56% 2452/4393 [3:22:55<2:09:32, 4.00s/it] + Training...: 56% 2453/4393 [3:23:01<2:29:31, 4.62s/it] + Training...: 56% 2454/4393 [3:23:07<2:42:44, 5.04s/it] + Training...: 56% 2455/4393 [3:23:13<2:51:51, 5.32s/it] + Training...: 56% 2456/4393 [3:23:19<2:57:11, 5.49s/it] + Training...: 56% 2457/4393 [3:23:25<3:00:58, 5.61s/it] + Training...: 56% 2458/4393 [3:23:31<3:03:11, 5.68s/it] + Training...: 56% 2459/4393 [3:23:36<3:04:53, 5.74s/it] + Training...: 56% 2460/4393 [3:23:42<3:04:45, 5.73s/it] + Training...: 56% 2461/4393 [3:23:48<3:05:13, 5.75s/it] + Training...: 56% 2462/4393 [3:23:54<3:04:58, 5.75s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:01:00<17:31:17, 21025.84s/it] + Training...: 56% 2462/4393 [3:24:00<3:04:58, 5.75s/it] + Training...: 56% 2463/4393 [3:24:00<3:12:22, 5.98s/it] + Training...: 56% 2464/4393 [3:24:06<3:08:33, 5.86s/it] + Training...: 56% 2465/4393 [3:24:11<3:06:34, 5.81s/it] + Training...: 56% 2466/4393 [3:24:17<3:04:36, 5.75s/it] + Training...: 56% 2467/4393 [3:24:23<3:03:03, 5.70s/it] + Training...: 56% 2468/4393 [3:24:28<3:00:45, 5.63s/it] + Training...: 56% 2469/4393 [3:24:34<2:59:06, 5.59s/it] + Training...: 56% 2470/4393 [3:24:39<2:57:26, 5.54s/it] + Training...: 56% 2471/4393 [3:24:44<2:56:08, 5.50s/it] + Training...: 56% 2472/4393 [3:24:50<2:54:44, 5.46s/it] + Training...: 56% 2473/4393 [3:24:55<2:53:56, 5.44s/it] + Training...: 56% 2474/4393 [3:25:00<2:52:13, 5.39s/it] + Training...: 56% 2475/4393 [3:25:06<2:51:15, 5.36s/it] + Training...: 56% 2476/4393 [3:25:11<2:50:43, 5.34s/it] + Training...: 56% 2477/4393 [3:25:16<2:50:14, 5.33s/it] + Training...: 56% 2478/4393 [3:25:22<2:48:45, 5.29s/it] + Training...: 56% 2479/4393 [3:25:27<2:46:26, 5.22s/it] + Training...: 56% 2480/4393 [3:25:32<2:44:07, 5.15s/it] + Training...: 56% 2481/4393 [3:25:37<2:42:44, 5.11s/it] + Training...: 56% 2482/4393 [3:25:41<2:39:56, 5.02s/it] + Training...: 57% 2483/4393 [3:25:46<2:37:43, 4.95s/it] + Training...: 57% 2484/4393 [3:25:51<2:34:49, 4.87s/it] + Training...: 57% 2485/4393 [3:25:55<2:31:49, 4.77s/it] + Training...: 57% 2486/4393 [3:26:00<2:28:04, 4.66s/it] + Training...: 57% 2487/4393 [3:26:04<2:24:38, 4.55s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:03:09<17:31:17, 21025.84s/it] + Training...: 57% 2487/4393 [3:26:09<2:24:38, 4.55s/it] + Training...: 57% 2488/4393 [3:26:09<2:26:13, 4.61s/it] + Training...: 57% 2489/4393 [3:26:13<2:19:32, 4.40s/it] + Training...: 57% 2490/4393 [3:26:16<2:11:53, 4.16s/it] + Training...: 57% 2491/4393 [3:26:20<2:06:42, 4.00s/it] + Training...: 57% 2492/4393 [3:26:23<2:00:43, 3.81s/it] + Training...: 57% 2493/4393 [3:26:27<1:54:12, 3.61s/it] + Training...: 57% 2494/4393 [3:26:29<1:46:56, 3.38s/it] + Training...: 57% 2495/4393 [3:26:32<1:39:28, 3.14s/it] + Training...: 57% 2496/4393 [3:26:34<1:31:26, 2.89s/it] + Training...: 57% 2497/4393 [3:26:36<1:23:24, 2.64s/it] + Training...: 57% 2498/4393 [3:26:38<1:14:42, 2.37s/it] + Training...: 57% 2499/4393 [3:26:39<1:06:02, 2.09s/it] + Training...: 57% 2500/4393 [3:26:41<56:53, 1.80s/it]  + Training...: 57% 2501/4393 [3:26:47<1:43:43, 3.29s/it] + Training...: 57% 2502/4393 [3:26:54<2:16:40, 4.34s/it] + Training...: 57% 2503/4393 [3:27:01<2:38:35, 5.03s/it] + Training...: 57% 2504/4393 [3:27:07<2:51:12, 5.44s/it] + Training...: 57% 2505/4393 [3:27:14<3:00:29, 5.74s/it] + Training...: 57% 2506/4393 [3:27:20<3:06:35, 5.93s/it] + Training...: 57% 2507/4393 [3:27:26<3:10:08, 6.05s/it] + Training...: 57% 2508/4393 [3:27:33<3:11:52, 6.11s/it] + Training...: 57% 2509/4393 [3:27:39<3:13:13, 6.15s/it] + Training...: 57% 2510/4393 [3:27:45<3:13:06, 6.15s/it] + Training...: 57% 2511/4393 [3:27:51<3:13:23, 6.17s/it] + Training...: 57% 2512/4393 [3:27:57<3:12:55, 6.15s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:05:04<17:31:17, 21025.84s/it] + Training...: 57% 2512/4393 [3:28:04<3:12:55, 6.15s/it] + Training...: 57% 2513/4393 [3:28:04<3:20:00, 6.38s/it] + Training...: 57% 2514/4393 [3:28:11<3:19:10, 6.36s/it] + Training...: 57% 2515/4393 [3:28:17<3:18:32, 6.34s/it] + Training...: 57% 2516/4393 [3:28:23<3:15:55, 6.26s/it] + Training...: 57% 2517/4393 [3:28:29<3:10:32, 6.09s/it] + Training...: 57% 2518/4393 [3:28:34<3:05:23, 5.93s/it] + Training...: 57% 2519/4393 [3:28:40<3:01:42, 5.82s/it] + Training...: 57% 2520/4393 [3:28:45<2:57:57, 5.70s/it] + Training...: 57% 2521/4393 [3:28:51<2:55:23, 5.62s/it] + Training...: 57% 2522/4393 [3:28:56<2:53:08, 5.55s/it] + Training...: 57% 2523/4393 [3:29:01<2:51:15, 5.50s/it] + Training...: 57% 2524/4393 [3:29:07<2:49:14, 5.43s/it] + Training...: 57% 2525/4393 [3:29:12<2:48:07, 5.40s/it] + Training...: 58% 2526/4393 [3:29:17<2:46:08, 5.34s/it] + Training...: 58% 2527/4393 [3:29:22<2:44:42, 5.30s/it] + Training...: 58% 2528/4393 [3:29:27<2:42:51, 5.24s/it] + Training...: 58% 2529/4393 [3:29:33<2:41:13, 5.19s/it] + Training...: 58% 2530/4393 [3:29:38<2:39:20, 5.13s/it] + Training...: 58% 2531/4393 [3:29:42<2:37:24, 5.07s/it] + Training...: 58% 2532/4393 [3:29:47<2:35:15, 5.01s/it] + Training...: 58% 2533/4393 [3:29:52<2:33:43, 4.96s/it] + Training...: 58% 2534/4393 [3:29:57<2:32:29, 4.92s/it] + Training...: 58% 2535/4393 [3:30:02<2:29:19, 4.82s/it] + Training...: 58% 2536/4393 [3:30:06<2:26:07, 4.72s/it] + Training...: 58% 2537/4393 [3:30:10<2:22:11, 4.60s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:07:15<17:31:17, 21025.84s/it] + Training...: 58% 2537/4393 [3:30:15<2:22:11, 4.60s/it] + Training...: 58% 2538/4393 [3:30:15<2:23:26, 4.64s/it] + Training...: 58% 2539/4393 [3:30:19<2:17:39, 4.45s/it] + Training...: 58% 2540/4393 [3:30:23<2:09:39, 4.20s/it] + Training...: 58% 2541/4393 [3:30:26<2:02:00, 3.95s/it] + Training...: 58% 2542/4393 [3:30:29<1:55:24, 3.74s/it] + Training...: 58% 2543/4393 [3:30:32<1:47:19, 3.48s/it] + Training...: 58% 2544/4393 [3:30:35<1:39:03, 3.21s/it] + Training...: 58% 2545/4393 [3:30:37<1:30:34, 2.94s/it] + Training...: 58% 2546/4393 [3:30:39<1:22:24, 2.68s/it] + Training...: 58% 2547/4393 [3:30:41<1:14:15, 2.41s/it] + Training...: 58% 2548/4393 [3:30:43<1:06:10, 2.15s/it] + Training...: 58% 2549/4393 [3:30:44<58:02, 1.89s/it]  + Training...: 58% 2550/4393 [3:30:45<49:59, 1.63s/it] + Training...: 58% 2551/4393 [3:30:51<1:32:26, 3.01s/it] + Training...: 58% 2552/4393 [3:30:57<2:02:30, 3.99s/it] + Training...: 58% 2553/4393 [3:31:04<2:22:29, 4.65s/it] + Training...: 58% 2554/4393 [3:31:10<2:34:53, 5.05s/it] + Training...: 58% 2555/4393 [3:31:16<2:43:20, 5.33s/it] + Training...: 58% 2556/4393 [3:31:21<2:48:54, 5.52s/it] + Training...: 58% 2557/4393 [3:31:27<2:52:56, 5.65s/it] + Training...: 58% 2558/4393 [3:31:33<2:54:20, 5.70s/it] + Training...: 58% 2559/4393 [3:31:39<2:55:30, 5.74s/it] + Training...: 58% 2560/4393 [3:31:45<2:55:51, 5.76s/it] + Training...: 58% 2561/4393 [3:31:51<2:56:14, 5.77s/it] + Training...: 58% 2562/4393 [3:31:56<2:55:46, 5.76s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:09:03<17:31:17, 21025.84s/it] + Training...: 58% 2562/4393 [3:32:03<2:55:46, 5.76s/it] + Training...: 58% 2563/4393 [3:32:03<3:03:52, 6.03s/it] + Training...: 58% 2564/4393 [3:32:09<3:01:10, 5.94s/it] + Training...: 58% 2565/4393 [3:32:14<2:58:37, 5.86s/it] + Training...: 58% 2566/4393 [3:32:20<2:57:37, 5.83s/it] + Training...: 58% 2567/4393 [3:32:26<2:57:14, 5.82s/it] + Training...: 58% 2568/4393 [3:32:32<2:54:11, 5.73s/it] + Training...: 58% 2569/4393 [3:32:37<2:51:43, 5.65s/it] + Training...: 59% 2570/4393 [3:32:42<2:49:55, 5.59s/it] + Training...: 59% 2571/4393 [3:32:48<2:47:49, 5.53s/it] + Training...: 59% 2572/4393 [3:32:53<2:45:51, 5.47s/it] + Training...: 59% 2573/4393 [3:32:59<2:45:03, 5.44s/it] + Training...: 59% 2574/4393 [3:33:04<2:43:25, 5.39s/it] + Training...: 59% 2575/4393 [3:33:09<2:42:10, 5.35s/it] + Training...: 59% 2576/4393 [3:33:14<2:40:56, 5.31s/it] + Training...: 59% 2577/4393 [3:33:19<2:39:31, 5.27s/it] + Training...: 59% 2578/4393 [3:33:25<2:38:01, 5.22s/it] + Training...: 59% 2579/4393 [3:33:30<2:36:39, 5.18s/it] + Training...: 59% 2580/4393 [3:33:35<2:34:52, 5.13s/it] + Training...: 59% 2581/4393 [3:33:40<2:33:36, 5.09s/it] + Training...: 59% 2582/4393 [3:33:45<2:31:24, 5.02s/it] + Training...: 59% 2583/4393 [3:33:49<2:28:47, 4.93s/it] + Training...: 59% 2584/4393 [3:33:54<2:26:46, 4.87s/it] + Training...: 59% 2585/4393 [3:33:59<2:25:32, 4.83s/it] + Training...: 59% 2586/4393 [3:34:03<2:21:38, 4.70s/it] + Training...: 59% 2587/4393 [3:34:07<2:17:30, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:11:12<17:31:17, 21025.84s/it] + Training...: 59% 2587/4393 [3:34:12<2:17:30, 4.57s/it] + Training...: 59% 2588/4393 [3:34:12<2:18:40, 4.61s/it] + Training...: 59% 2589/4393 [3:34:16<2:11:16, 4.37s/it] + Training...: 59% 2590/4393 [3:34:19<2:03:52, 4.12s/it] + Training...: 59% 2591/4393 [3:34:23<1:56:30, 3.88s/it] + Training...: 59% 2592/4393 [3:34:26<1:49:25, 3.65s/it] + Training...: 59% 2593/4393 [3:34:29<1:41:51, 3.40s/it] + Training...: 59% 2594/4393 [3:34:31<1:33:57, 3.13s/it] + Training...: 59% 2595/4393 [3:34:33<1:26:09, 2.88s/it] + Training...: 59% 2596/4393 [3:34:35<1:18:07, 2.61s/it] + Training...: 59% 2597/4393 [3:34:37<1:10:08, 2.34s/it] + Training...: 59% 2598/4393 [3:34:39<1:02:32, 2.09s/it] + Training...: 59% 2599/4393 [3:34:40<54:52, 1.84s/it]  + Training...: 59% 2600/4393 [3:34:41<47:19, 1.58s/it] + Training...: 59% 2601/4393 [3:34:47<1:28:41, 2.97s/it] + Training...: 59% 2602/4393 [3:34:53<1:58:22, 3.97s/it] + Training...: 59% 2603/4393 [3:35:00<2:17:34, 4.61s/it] + Training...: 59% 2604/4393 [3:35:06<2:30:06, 5.03s/it] + Training...: 59% 2605/4393 [3:35:12<2:38:38, 5.32s/it] + Training...: 59% 2606/4393 [3:35:17<2:44:12, 5.51s/it] + Training...: 59% 2607/4393 [3:35:23<2:47:51, 5.64s/it] + Training...: 59% 2608/4393 [3:35:30<2:52:02, 5.78s/it] + Training...: 59% 2609/4393 [3:35:35<2:52:46, 5.81s/it] + Training...: 59% 2610/4393 [3:35:41<2:52:26, 5.80s/it] + Training...: 59% 2611/4393 [3:35:47<2:52:01, 5.79s/it] + Training...: 59% 2612/4393 [3:35:53<2:51:07, 5.77s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:12:59<17:31:17, 21025.84s/it] + Training...: 59% 2612/4393 [3:35:59<2:51:07, 5.77s/it] + Training...: 59% 2613/4393 [3:35:59<2:57:41, 5.99s/it] + Training...: 60% 2614/4393 [3:36:05<2:54:53, 5.90s/it] + Training...: 60% 2615/4393 [3:36:11<2:53:16, 5.85s/it] + Training...: 60% 2616/4393 [3:36:16<2:51:11, 5.78s/it] + Training...: 60% 2617/4393 [3:36:22<2:50:17, 5.75s/it] + Training...: 60% 2618/4393 [3:36:27<2:48:42, 5.70s/it] + Training...: 60% 2619/4393 [3:36:33<2:47:22, 5.66s/it] + Training...: 60% 2620/4393 [3:36:39<2:46:12, 5.62s/it] + Training...: 60% 2621/4393 [3:36:44<2:44:37, 5.57s/it] + Training...: 60% 2622/4393 [3:36:49<2:43:04, 5.52s/it] + Training...: 60% 2623/4393 [3:36:55<2:40:57, 5.46s/it] + Training...: 60% 2624/4393 [3:37:00<2:42:12, 5.50s/it] + Training...: 60% 2625/4393 [3:37:06<2:40:10, 5.44s/it] + Training...: 60% 2626/4393 [3:37:11<2:37:59, 5.36s/it] + Training...: 60% 2627/4393 [3:37:16<2:36:29, 5.32s/it] + Training...: 60% 2628/4393 [3:37:21<2:34:37, 5.26s/it] + Training...: 60% 2629/4393 [3:37:26<2:33:14, 5.21s/it] + Training...: 60% 2630/4393 [3:37:31<2:31:07, 5.14s/it] + Training...: 60% 2631/4393 [3:37:36<2:29:18, 5.08s/it] + Training...: 60% 2632/4393 [3:37:41<2:27:35, 5.03s/it] + Training...: 60% 2633/4393 [3:37:46<2:25:48, 4.97s/it] + Training...: 60% 2634/4393 [3:37:51<2:23:13, 4.89s/it] + Training...: 60% 2635/4393 [3:37:55<2:20:26, 4.79s/it] + Training...: 60% 2636/4393 [3:38:00<2:17:18, 4.69s/it] + Training...: 60% 2637/4393 [3:38:04<2:13:11, 4.55s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:15:09<17:31:17, 21025.84s/it] + Training...: 60% 2637/4393 [3:38:09<2:13:11, 4.55s/it] + Training...: 60% 2638/4393 [3:38:09<2:14:38, 4.60s/it] + Training...: 60% 2639/4393 [3:38:12<2:08:11, 4.38s/it] + Training...: 60% 2640/4393 [3:38:16<2:01:21, 4.15s/it] + Training...: 60% 2641/4393 [3:38:19<1:54:29, 3.92s/it] + Training...: 60% 2642/4393 [3:38:23<1:47:11, 3.67s/it] + Training...: 60% 2643/4393 [3:38:25<1:39:52, 3.42s/it] + Training...: 60% 2644/4393 [3:38:28<1:32:51, 3.19s/it] + Training...: 60% 2645/4393 [3:38:30<1:25:22, 2.93s/it] + Training...: 60% 2646/4393 [3:38:32<1:17:59, 2.68s/it] + Training...: 60% 2647/4393 [3:38:34<1:10:50, 2.43s/it] + Training...: 60% 2648/4393 [3:38:36<1:03:24, 2.18s/it] + Training...: 60% 2649/4393 [3:38:37<55:43, 1.92s/it]  + Training...: 60% 2650/4393 [3:38:38<47:55, 1.65s/it] + Training...: 60% 2651/4393 [3:38:45<1:28:26, 3.05s/it] + Training...: 60% 2652/4393 [3:38:51<1:56:29, 4.01s/it] + Training...: 60% 2653/4393 [3:38:57<2:14:53, 4.65s/it] + Training...: 60% 2654/4393 [3:39:03<2:26:41, 5.06s/it] + Training...: 60% 2655/4393 [3:39:09<2:35:56, 5.38s/it] + Training...: 60% 2656/4393 [3:39:15<2:40:16, 5.54s/it] + Training...: 60% 2657/4393 [3:39:21<2:42:54, 5.63s/it] + Training...: 61% 2658/4393 [3:39:27<2:44:52, 5.70s/it] + Training...: 61% 2659/4393 [3:39:33<2:46:19, 5.76s/it] + Training...: 61% 2660/4393 [3:39:38<2:46:29, 5.76s/it] + Training...: 61% 2661/4393 [3:39:44<2:46:36, 5.77s/it] + Training...: 61% 2662/4393 [3:39:50<2:46:00, 5.75s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:16:56<17:31:17, 21025.84s/it] + Training...: 61% 2662/4393 [3:39:56<2:46:00, 5.75s/it] + Training...: 61% 2663/4393 [3:39:56<2:52:54, 6.00s/it] + Training...: 61% 2664/4393 [3:40:02<2:52:29, 5.99s/it] + Training...: 61% 2665/4393 [3:40:08<2:50:40, 5.93s/it] + Training...: 61% 2666/4393 [3:40:14<2:47:57, 5.84s/it] + Training...: 61% 2667/4393 [3:40:19<2:46:08, 5.78s/it] + Training...: 61% 2668/4393 [3:40:25<2:44:53, 5.74s/it] + Training...: 61% 2669/4393 [3:40:31<2:43:16, 5.68s/it] + Training...: 61% 2670/4393 [3:40:36<2:41:23, 5.62s/it] + Training...: 61% 2671/4393 [3:40:42<2:39:58, 5.57s/it] + Training...: 61% 2672/4393 [3:40:47<2:37:46, 5.50s/it] + Training...: 61% 2673/4393 [3:40:52<2:36:23, 5.46s/it] + Training...: 61% 2674/4393 [3:40:58<2:35:06, 5.41s/it] + Training...: 61% 2675/4393 [3:41:03<2:34:15, 5.39s/it] + Training...: 61% 2676/4393 [3:41:08<2:32:20, 5.32s/it] + Training...: 61% 2677/4393 [3:41:13<2:31:09, 5.29s/it] + Training...: 61% 2678/4393 [3:41:18<2:29:15, 5.22s/it] + Training...: 61% 2679/4393 [3:41:24<2:28:24, 5.20s/it] + Training...: 61% 2680/4393 [3:41:28<2:26:27, 5.13s/it] + Training...: 61% 2681/4393 [3:41:33<2:24:42, 5.07s/it] + Training...: 61% 2682/4393 [3:41:38<2:24:11, 5.06s/it] + Training...: 61% 2683/4393 [3:41:43<2:23:03, 5.02s/it] + Training...: 61% 2684/4393 [3:41:48<2:19:55, 4.91s/it] + Training...: 61% 2685/4393 [3:41:53<2:17:20, 4.82s/it] + Training...: 61% 2686/4393 [3:41:57<2:14:15, 4.72s/it] + Training...: 61% 2687/4393 [3:42:01<2:10:52, 4.60s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:19:06<17:31:17, 21025.84s/it] + Training...: 61% 2687/4393 [3:42:06<2:10:52, 4.60s/it] + Training...: 61% 2688/4393 [3:42:06<2:12:03, 4.65s/it] + Training...: 61% 2689/4393 [3:42:10<2:06:17, 4.45s/it] + Training...: 61% 2690/4393 [3:42:14<1:59:49, 4.22s/it] + Training...: 61% 2691/4393 [3:42:17<1:53:21, 4.00s/it] + Training...: 61% 2692/4393 [3:42:21<1:46:21, 3.75s/it] + Training...: 61% 2693/4393 [3:42:24<1:39:43, 3.52s/it] + Training...: 61% 2694/4393 [3:42:26<1:32:39, 3.27s/it] + Training...: 61% 2695/4393 [3:42:29<1:25:00, 3.00s/it] + Training...: 61% 2696/4393 [3:42:31<1:17:36, 2.74s/it] + Training...: 61% 2697/4393 [3:42:33<1:10:26, 2.49s/it] + Training...: 61% 2698/4393 [3:42:34<1:03:01, 2.23s/it] + Training...: 61% 2699/4393 [3:42:36<55:39, 1.97s/it]  + Training...: 61% 2700/4393 [3:42:37<47:57, 1.70s/it] + Training...: 61% 2701/4393 [3:42:43<1:26:56, 3.08s/it] + Training...: 62% 2702/4393 [3:42:49<1:53:49, 4.04s/it] + Training...: 62% 2703/4393 [3:42:55<2:11:06, 4.65s/it] + Training...: 62% 2704/4393 [3:43:01<2:22:34, 5.06s/it] + Training...: 62% 2705/4393 [3:43:07<2:31:20, 5.38s/it] + Training...: 62% 2706/4393 [3:43:13<2:36:01, 5.55s/it] + Training...: 62% 2707/4393 [3:43:19<2:38:51, 5.65s/it] + Training...: 62% 2708/4393 [3:43:25<2:41:57, 5.77s/it] + Training...: 62% 2709/4393 [3:43:31<2:44:22, 5.86s/it] + Training...: 62% 2710/4393 [3:43:37<2:43:55, 5.84s/it] + Training...: 62% 2711/4393 [3:43:43<2:43:14, 5.82s/it] + Training...: 62% 2712/4393 [3:43:49<2:41:58, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:20:55<17:31:17, 21025.84s/it] + Training...: 62% 2712/4393 [3:43:55<2:41:58, 5.78s/it] + Training...: 62% 2713/4393 [3:43:55<2:48:19, 6.01s/it] + Training...: 62% 2714/4393 [3:44:01<2:45:13, 5.90s/it] + Training...: 62% 2715/4393 [3:44:07<2:42:46, 5.82s/it] + Training...: 62% 2716/4393 [3:44:12<2:41:19, 5.77s/it] + Training...: 62% 2717/4393 [3:44:18<2:39:55, 5.73s/it] + Training...: 62% 2718/4393 [3:44:23<2:38:23, 5.67s/it] + Training...: 62% 2719/4393 [3:44:29<2:36:57, 5.63s/it] + Training...: 62% 2720/4393 [3:44:34<2:35:02, 5.56s/it] + Training...: 62% 2721/4393 [3:44:40<2:33:36, 5.51s/it] + Training...: 62% 2722/4393 [3:44:45<2:31:54, 5.45s/it] + Training...: 62% 2723/4393 [3:44:50<2:31:08, 5.43s/it] + Training...: 62% 2724/4393 [3:44:56<2:30:37, 5.41s/it] + Training...: 62% 2725/4393 [3:45:01<2:30:20, 5.41s/it] + Training...: 62% 2726/4393 [3:45:06<2:28:25, 5.34s/it] + Training...: 62% 2727/4393 [3:45:11<2:26:38, 5.28s/it] + Training...: 62% 2728/4393 [3:45:17<2:24:38, 5.21s/it] + Training...: 62% 2729/4393 [3:45:22<2:22:57, 5.15s/it] + Training...: 62% 2730/4393 [3:45:26<2:20:53, 5.08s/it] + Training...: 62% 2731/4393 [3:45:31<2:19:36, 5.04s/it] + Training...: 62% 2732/4393 [3:45:36<2:17:27, 4.97s/it] + Training...: 62% 2733/4393 [3:45:41<2:15:35, 4.90s/it] + Training...: 62% 2734/4393 [3:45:46<2:13:01, 4.81s/it] + Training...: 62% 2735/4393 [3:45:50<2:10:47, 4.73s/it] + Training...: 62% 2736/4393 [3:45:54<2:07:32, 4.62s/it] + Training...: 62% 2737/4393 [3:45:59<2:04:15, 4.50s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:23:03<17:31:17, 21025.84s/it] + Training...: 62% 2737/4393 [3:46:03<2:04:15, 4.50s/it] + Training...: 62% 2738/4393 [3:46:03<2:05:44, 4.56s/it] + Training...: 62% 2739/4393 [3:46:07<2:00:18, 4.36s/it] + Training...: 62% 2740/4393 [3:46:11<1:54:17, 4.15s/it] + Training...: 62% 2741/4393 [3:46:14<1:47:43, 3.91s/it] + Training...: 62% 2742/4393 [3:46:17<1:40:30, 3.65s/it] + Training...: 62% 2743/4393 [3:46:20<1:33:36, 3.40s/it] + Training...: 62% 2744/4393 [3:46:23<1:26:59, 3.17s/it] + Training...: 62% 2745/4393 [3:46:25<1:20:19, 2.92s/it] + Training...: 63% 2746/4393 [3:46:27<1:13:41, 2.68s/it] + Training...: 63% 2747/4393 [3:46:29<1:07:07, 2.45s/it] + Training...: 63% 2748/4393 [3:46:31<1:00:02, 2.19s/it] + Training...: 63% 2749/4393 [3:46:32<52:59, 1.93s/it]  + Training...: 63% 2750/4393 [3:46:33<45:38, 1.67s/it] + Training...: 63% 2751/4393 [3:46:40<1:25:25, 3.12s/it] + Training...: 63% 2752/4393 [3:46:46<1:51:51, 4.09s/it] + Training...: 63% 2753/4393 [3:46:52<2:08:16, 4.69s/it] + Training...: 63% 2754/4393 [3:46:58<2:18:56, 5.09s/it] + Training...: 63% 2755/4393 [3:47:04<2:26:15, 5.36s/it] + Training...: 63% 2756/4393 [3:47:10<2:31:24, 5.55s/it] + Training...: 63% 2757/4393 [3:47:16<2:35:21, 5.70s/it] + Training...: 63% 2758/4393 [3:47:22<2:37:14, 5.77s/it] + Training...: 63% 2759/4393 [3:47:28<2:37:48, 5.79s/it] + Training...: 63% 2760/4393 [3:47:34<2:37:44, 5.80s/it] + Training...: 63% 2761/4393 [3:47:39<2:37:28, 5.79s/it] + Training...: 63% 2762/4393 [3:47:45<2:37:21, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:24:52<17:31:17, 21025.84s/it] + Training...: 63% 2762/4393 [3:47:52<2:37:21, 5.79s/it] + Training...: 63% 2763/4393 [3:47:52<2:43:49, 6.03s/it] + Training...: 63% 2764/4393 [3:47:58<2:41:58, 5.97s/it] + Training...: 63% 2765/4393 [3:48:03<2:39:23, 5.87s/it] + Training...: 63% 2766/4393 [3:48:09<2:36:59, 5.79s/it] + Training...: 63% 2767/4393 [3:48:15<2:35:39, 5.74s/it] + Training...: 63% 2768/4393 [3:48:20<2:34:05, 5.69s/it] + Training...: 63% 2769/4393 [3:48:26<2:33:05, 5.66s/it] + Training...: 63% 2770/4393 [3:48:31<2:31:26, 5.60s/it] + Training...: 63% 2771/4393 [3:48:37<2:29:51, 5.54s/it] + Training...: 63% 2772/4393 [3:48:42<2:28:05, 5.48s/it] + Training...: 63% 2773/4393 [3:48:47<2:26:43, 5.43s/it] + Training...: 63% 2774/4393 [3:48:53<2:25:25, 5.39s/it] + Training...: 63% 2775/4393 [3:48:58<2:25:30, 5.40s/it] + Training...: 63% 2776/4393 [3:49:03<2:24:15, 5.35s/it] + Training...: 63% 2777/4393 [3:49:08<2:23:36, 5.33s/it] + Training...: 63% 2778/4393 [3:49:14<2:22:01, 5.28s/it] + Training...: 63% 2779/4393 [3:49:19<2:20:32, 5.22s/it] + Training...: 63% 2780/4393 [3:49:24<2:18:46, 5.16s/it] + Training...: 63% 2781/4393 [3:49:29<2:17:11, 5.11s/it] + Training...: 63% 2782/4393 [3:49:34<2:14:55, 5.02s/it] + Training...: 63% 2783/4393 [3:49:38<2:12:47, 4.95s/it] + Training...: 63% 2784/4393 [3:49:43<2:10:39, 4.87s/it] + Training...: 63% 2785/4393 [3:49:48<2:08:49, 4.81s/it] + Training...: 63% 2786/4393 [3:49:52<2:06:11, 4.71s/it] + Training...: 63% 2787/4393 [3:49:56<2:02:46, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:27:01<17:31:17, 21025.84s/it] + Training...: 63% 2787/4393 [3:50:01<2:02:46, 4.59s/it] + Training...: 63% 2788/4393 [3:50:01<2:03:58, 4.63s/it] + Training...: 63% 2789/4393 [3:50:05<1:58:09, 4.42s/it] + Training...: 64% 2790/4393 [3:50:09<1:52:09, 4.20s/it] + Training...: 64% 2791/4393 [3:50:12<1:46:34, 3.99s/it] + Training...: 64% 2792/4393 [3:50:16<1:40:21, 3.76s/it] + Training...: 64% 2793/4393 [3:50:18<1:33:48, 3.52s/it] + Training...: 64% 2794/4393 [3:50:21<1:26:48, 3.26s/it] + Training...: 64% 2795/4393 [3:50:24<1:19:56, 3.00s/it] + Training...: 64% 2796/4393 [3:50:26<1:12:47, 2.73s/it] + Training...: 64% 2797/4393 [3:50:28<1:05:42, 2.47s/it] + Training...: 64% 2798/4393 [3:50:29<58:36, 2.20s/it]  + Training...: 64% 2799/4393 [3:50:30<51:28, 1.94s/it] + Training...: 64% 2800/4393 [3:50:31<44:03, 1.66s/it] + Training...: 64% 2801/4393 [3:50:38<1:20:49, 3.05s/it] + Training...: 64% 2802/4393 [3:50:44<1:45:56, 4.00s/it] + Training...: 64% 2803/4393 [3:50:50<2:04:43, 4.71s/it] + Training...: 64% 2804/4393 [3:50:57<2:18:10, 5.22s/it] + Training...: 64% 2805/4393 [3:51:03<2:24:15, 5.45s/it] + Training...: 64% 2806/4393 [3:51:09<2:28:09, 5.60s/it] + Training...: 64% 2807/4393 [3:51:15<2:30:30, 5.69s/it] + Training...: 64% 2808/4393 [3:51:21<2:32:59, 5.79s/it] + Training...: 64% 2809/4393 [3:51:26<2:33:24, 5.81s/it] + Training...: 64% 2810/4393 [3:51:32<2:34:49, 5.87s/it] + Training...: 64% 2811/4393 [3:51:38<2:34:20, 5.85s/it] + Training...: 64% 2812/4393 [3:51:44<2:33:38, 5.83s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:28:50<17:31:17, 21025.84s/it] + Training...: 64% 2812/4393 [3:51:51<2:33:38, 5.83s/it] + Training...: 64% 2813/4393 [3:51:51<2:39:08, 6.04s/it] + Training...: 64% 2814/4393 [3:51:56<2:37:02, 5.97s/it] + Training...: 64% 2815/4393 [3:52:02<2:34:30, 5.88s/it] + Training...: 64% 2816/4393 [3:52:08<2:32:41, 5.81s/it] + Training...: 64% 2817/4393 [3:52:13<2:31:25, 5.76s/it] + Training...: 64% 2818/4393 [3:52:19<2:30:32, 5.73s/it] + Training...: 64% 2819/4393 [3:52:25<2:30:53, 5.75s/it] + Training...: 64% 2820/4393 [3:52:30<2:28:44, 5.67s/it] + Training...: 64% 2821/4393 [3:52:36<2:26:46, 5.60s/it] + Training...: 64% 2822/4393 [3:52:41<2:24:51, 5.53s/it] + Training...: 64% 2823/4393 [3:52:46<2:23:11, 5.47s/it] + Training...: 64% 2824/4393 [3:52:52<2:21:49, 5.42s/it] + Training...: 64% 2825/4393 [3:52:57<2:20:59, 5.39s/it] + Training...: 64% 2826/4393 [3:53:02<2:19:16, 5.33s/it] + Training...: 64% 2827/4393 [3:53:07<2:18:26, 5.30s/it] + Training...: 64% 2828/4393 [3:53:13<2:16:56, 5.25s/it] + Training...: 64% 2829/4393 [3:53:18<2:15:31, 5.20s/it] + Training...: 64% 2830/4393 [3:53:23<2:13:32, 5.13s/it] + Training...: 64% 2831/4393 [3:53:28<2:12:01, 5.07s/it] + Training...: 64% 2832/4393 [3:53:32<2:10:10, 5.00s/it] + Training...: 64% 2833/4393 [3:53:37<2:08:27, 4.94s/it] + Training...: 65% 2834/4393 [3:53:42<2:06:16, 4.86s/it] + Training...: 65% 2835/4393 [3:53:46<2:03:52, 4.77s/it] + Training...: 65% 2836/4393 [3:53:51<2:02:09, 4.71s/it] + Training...: 65% 2837/4393 [3:53:55<1:59:52, 4.62s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:31:00<17:31:17, 21025.84s/it] + Training...: 65% 2837/4393 [3:54:00<1:59:52, 4.62s/it] + Training...: 65% 2838/4393 [3:54:00<2:00:23, 4.65s/it] + Training...: 65% 2839/4393 [3:54:04<1:54:31, 4.42s/it] + Training...: 65% 2840/4393 [3:54:08<1:48:07, 4.18s/it] + Training...: 65% 2841/4393 [3:54:11<1:42:34, 3.97s/it] + Training...: 65% 2842/4393 [3:54:14<1:36:31, 3.73s/it] + Training...: 65% 2843/4393 [3:54:17<1:30:31, 3.50s/it] + Training...: 65% 2844/4393 [3:54:20<1:24:33, 3.28s/it] + Training...: 65% 2845/4393 [3:54:22<1:17:50, 3.02s/it] + Training...: 65% 2846/4393 [3:54:25<1:11:17, 2.76s/it] + Training...: 65% 2847/4393 [3:54:26<1:03:59, 2.48s/it] + Training...: 65% 2848/4393 [3:54:28<57:05, 2.22s/it]  + Training...: 65% 2849/4393 [3:54:29<50:16, 1.95s/it] + Training...: 65% 2850/4393 [3:54:30<42:59, 1.67s/it] + Training...: 65% 2851/4393 [3:54:37<1:18:30, 3.05s/it] + Training...: 65% 2852/4393 [3:54:43<1:42:49, 4.00s/it] + Training...: 65% 2853/4393 [3:54:49<1:58:48, 4.63s/it] + Training...: 65% 2854/4393 [3:54:55<2:09:59, 5.07s/it] + Training...: 65% 2855/4393 [3:55:01<2:17:08, 5.35s/it] + Training...: 65% 2856/4393 [3:55:07<2:22:26, 5.56s/it] + Training...: 65% 2857/4393 [3:55:13<2:25:53, 5.70s/it] + Training...: 65% 2858/4393 [3:55:19<2:27:25, 5.76s/it] + Training...: 65% 2859/4393 [3:55:25<2:30:18, 5.88s/it] + Training...: 65% 2860/4393 [3:55:31<2:29:51, 5.87s/it] + Training...: 65% 2861/4393 [3:55:37<2:29:12, 5.84s/it] + Training...: 65% 2862/4393 [3:55:43<2:27:57, 5.80s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:32:49<17:31:17, 21025.84s/it] + Training...: 65% 2862/4393 [3:55:49<2:27:57, 5.80s/it] + Training...: 65% 2863/4393 [3:55:49<2:33:41, 6.03s/it] + Training...: 65% 2864/4393 [3:55:55<2:31:27, 5.94s/it] + Training...: 65% 2865/4393 [3:56:01<2:29:33, 5.87s/it] + Training...: 65% 2866/4393 [3:56:06<2:27:35, 5.80s/it] + Training...: 65% 2867/4393 [3:56:12<2:26:02, 5.74s/it] + Training...: 65% 2868/4393 [3:56:17<2:24:22, 5.68s/it] + Training...: 65% 2869/4393 [3:56:23<2:23:30, 5.65s/it] + Training...: 65% 2870/4393 [3:56:28<2:22:50, 5.63s/it] + Training...: 65% 2871/4393 [3:56:34<2:21:58, 5.60s/it] + Training...: 65% 2872/4393 [3:56:39<2:20:00, 5.52s/it] + Training...: 65% 2873/4393 [3:56:45<2:18:45, 5.48s/it] + Training...: 65% 2874/4393 [3:56:50<2:17:11, 5.42s/it] + Training...: 65% 2875/4393 [3:56:55<2:15:55, 5.37s/it] + Training...: 65% 2876/4393 [3:57:00<2:14:38, 5.33s/it] + Training...: 65% 2877/4393 [3:57:06<2:15:11, 5.35s/it] + Training...: 66% 2878/4393 [3:57:11<2:13:25, 5.28s/it] + Training...: 66% 2879/4393 [3:57:16<2:11:51, 5.23s/it] + Training...: 66% 2880/4393 [3:57:21<2:09:42, 5.14s/it] + Training...: 66% 2881/4393 [3:57:26<2:07:42, 5.07s/it] + Training...: 66% 2882/4393 [3:57:31<2:05:18, 4.98s/it] + Training...: 66% 2883/4393 [3:57:35<2:03:17, 4.90s/it] + Training...: 66% 2884/4393 [3:57:40<2:01:17, 4.82s/it] + Training...: 66% 2885/4393 [3:57:45<1:59:07, 4.74s/it] + Training...: 66% 2886/4393 [3:57:49<1:56:08, 4.62s/it] + Training...: 66% 2887/4393 [3:57:53<1:53:16, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:34:58<17:31:17, 21025.84s/it] + Training...: 66% 2887/4393 [3:57:58<1:53:16, 4.51s/it] + Training...: 66% 2888/4393 [3:57:58<1:53:49, 4.54s/it] + Training...: 66% 2889/4393 [3:58:02<1:48:19, 4.32s/it] + Training...: 66% 2890/4393 [3:58:05<1:42:47, 4.10s/it] + Training...: 66% 2891/4393 [3:58:09<1:36:43, 3.86s/it] + Training...: 66% 2892/4393 [3:58:12<1:30:43, 3.63s/it] + Training...: 66% 2893/4393 [3:58:14<1:25:04, 3.40s/it] + Training...: 66% 2894/4393 [3:58:17<1:18:10, 3.13s/it] + Training...: 66% 2895/4393 [3:58:19<1:12:00, 2.88s/it] + Training...: 66% 2896/4393 [3:58:21<1:05:30, 2.63s/it] + Training...: 66% 2897/4393 [3:58:23<58:29, 2.35s/it]  + Training...: 66% 2898/4393 [3:58:24<51:49, 2.08s/it] + Training...: 66% 2899/4393 [3:58:26<45:45, 1.84s/it] + Training...: 66% 2900/4393 [3:58:27<39:46, 1.60s/it] + Training...: 66% 2901/4393 [3:58:33<1:14:27, 2.99s/it] + Training...: 66% 2902/4393 [3:58:39<1:38:42, 3.97s/it] + Training...: 66% 2903/4393 [3:58:45<1:54:34, 4.61s/it] + Training...: 66% 2904/4393 [3:58:51<2:05:22, 5.05s/it] + Training...: 66% 2905/4393 [3:58:57<2:12:23, 5.34s/it] + Training...: 66% 2906/4393 [3:59:03<2:16:49, 5.52s/it] + Training...: 66% 2907/4393 [3:59:09<2:20:15, 5.66s/it] + Training...: 66% 2908/4393 [3:59:15<2:22:13, 5.75s/it] + Training...: 66% 2909/4393 [3:59:21<2:23:01, 5.78s/it] + Training...: 66% 2910/4393 [3:59:27<2:23:23, 5.80s/it] + Training...: 66% 2911/4393 [3:59:33<2:25:03, 5.87s/it] + Training...: 66% 2912/4393 [3:59:39<2:25:15, 5.88s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:36:46<17:31:17, 21025.84s/it] + Training...: 66% 2912/4393 [3:59:46<2:25:15, 5.88s/it] + Training...: 66% 2913/4393 [3:59:46<2:30:23, 6.10s/it] + Training...: 66% 2914/4393 [3:59:51<2:27:17, 5.98s/it] + Training...: 66% 2915/4393 [3:59:57<2:25:03, 5.89s/it] + Training...: 66% 2916/4393 [4:00:03<2:22:52, 5.80s/it] + Training...: 66% 2917/4393 [4:00:08<2:21:25, 5.75s/it] + Training...: 66% 2918/4393 [4:00:14<2:19:41, 5.68s/it] + Training...: 66% 2919/4393 [4:00:19<2:18:15, 5.63s/it] + Training...: 66% 2920/4393 [4:00:25<2:16:59, 5.58s/it] + Training...: 66% 2921/4393 [4:00:30<2:16:19, 5.56s/it] + Training...: 67% 2922/4393 [4:00:36<2:14:40, 5.49s/it] + Training...: 67% 2923/4393 [4:00:41<2:13:16, 5.44s/it] + Training...: 67% 2924/4393 [4:00:46<2:11:32, 5.37s/it] + Training...: 67% 2925/4393 [4:00:51<2:10:34, 5.34s/it] + Training...: 67% 2926/4393 [4:00:57<2:09:31, 5.30s/it] + Training...: 67% 2927/4393 [4:01:02<2:09:06, 5.28s/it] + Training...: 67% 2928/4393 [4:01:07<2:08:07, 5.25s/it] + Training...: 67% 2929/4393 [4:01:12<2:08:06, 5.25s/it] + Training...: 67% 2930/4393 [4:01:17<2:06:57, 5.21s/it] + Training...: 67% 2931/4393 [4:01:22<2:04:34, 5.11s/it] + Training...: 67% 2932/4393 [4:01:27<2:02:32, 5.03s/it] + Training...: 67% 2933/4393 [4:01:32<2:00:48, 4.96s/it] + Training...: 67% 2934/4393 [4:01:37<1:58:42, 4.88s/it] + Training...: 67% 2935/4393 [4:01:41<1:56:05, 4.78s/it] + Training...: 67% 2936/4393 [4:01:46<1:53:16, 4.67s/it] + Training...: 67% 2937/4393 [4:01:50<1:50:08, 4.54s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:38:54<17:31:17, 21025.84s/it] + Training...: 67% 2937/4393 [4:01:54<1:50:08, 4.54s/it] + Training...: 67% 2938/4393 [4:01:54<1:51:09, 4.58s/it] + Training...: 67% 2939/4393 [4:01:58<1:46:00, 4.37s/it] + Training...: 67% 2940/4393 [4:02:02<1:40:18, 4.14s/it] + Training...: 67% 2941/4393 [4:02:05<1:34:41, 3.91s/it] + Training...: 67% 2942/4393 [4:02:08<1:28:49, 3.67s/it] + Training...: 67% 2943/4393 [4:02:11<1:23:15, 3.45s/it] + Training...: 67% 2944/4393 [4:02:14<1:17:11, 3.20s/it] + Training...: 67% 2945/4393 [4:02:16<1:10:43, 2.93s/it] + Training...: 67% 2946/4393 [4:02:18<1:04:28, 2.67s/it] + Training...: 67% 2947/4393 [4:02:20<58:19, 2.42s/it]  + Training...: 67% 2948/4393 [4:02:22<51:58, 2.16s/it] + Training...: 67% 2949/4393 [4:02:23<45:17, 1.88s/it] + Training...: 67% 2950/4393 [4:02:24<38:45, 1.61s/it] + Training...: 67% 2951/4393 [4:02:30<1:13:02, 3.04s/it] + Training...: 67% 2952/4393 [4:02:37<1:37:30, 4.06s/it] + Training...: 67% 2953/4393 [4:02:43<1:52:39, 4.69s/it] + Training...: 67% 2954/4393 [4:02:49<2:01:41, 5.07s/it] + Training...: 67% 2955/4393 [4:02:55<2:08:33, 5.36s/it] + Training...: 67% 2956/4393 [4:03:01<2:12:13, 5.52s/it] + Training...: 67% 2957/4393 [4:03:07<2:14:48, 5.63s/it] + Training...: 67% 2958/4393 [4:03:13<2:17:07, 5.73s/it] + Training...: 67% 2959/4393 [4:03:19<2:18:17, 5.79s/it] + Training...: 67% 2960/4393 [4:03:24<2:18:56, 5.82s/it] + Training...: 67% 2961/4393 [4:03:30<2:19:04, 5.83s/it] + Training...: 67% 2962/4393 [4:03:36<2:18:42, 5.82s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:40:43<17:31:17, 21025.84s/it] + Training...: 67% 2962/4393 [4:03:43<2:18:42, 5.82s/it] + Training...: 67% 2963/4393 [4:03:43<2:25:55, 6.12s/it] + Training...: 67% 2964/4393 [4:03:49<2:22:54, 6.00s/it] + Training...: 67% 2965/4393 [4:03:54<2:20:47, 5.92s/it] + Training...: 68% 2966/4393 [4:04:00<2:19:10, 5.85s/it] + Training...: 68% 2967/4393 [4:04:06<2:17:27, 5.78s/it] + Training...: 68% 2968/4393 [4:04:11<2:15:39, 5.71s/it] + Training...: 68% 2969/4393 [4:04:17<2:14:04, 5.65s/it] + Training...: 68% 2970/4393 [4:04:22<2:12:35, 5.59s/it] + Training...: 68% 2971/4393 [4:04:28<2:11:46, 5.56s/it] + Training...: 68% 2972/4393 [4:04:33<2:11:19, 5.55s/it] + Training...: 68% 2973/4393 [4:04:39<2:10:03, 5.50s/it] + Training...: 68% 2974/4393 [4:04:44<2:08:32, 5.44s/it] + Training...: 68% 2975/4393 [4:04:49<2:07:29, 5.39s/it] + Training...: 68% 2976/4393 [4:04:54<2:06:12, 5.34s/it] + Training...: 68% 2977/4393 [4:05:00<2:05:01, 5.30s/it] + Training...: 68% 2978/4393 [4:05:05<2:03:41, 5.24s/it] + Training...: 68% 2979/4393 [4:05:10<2:02:13, 5.19s/it] + Training...: 68% 2980/4393 [4:05:15<2:01:37, 5.16s/it] + Training...: 68% 2981/4393 [4:05:20<2:01:06, 5.15s/it] + Training...: 68% 2982/4393 [4:05:25<1:58:59, 5.06s/it] + Training...: 68% 2983/4393 [4:05:30<1:56:43, 4.97s/it] + Training...: 68% 2984/4393 [4:05:34<1:54:17, 4.87s/it] + Training...: 68% 2985/4393 [4:05:39<1:52:16, 4.78s/it] + Training...: 68% 2986/4393 [4:05:43<1:49:26, 4.67s/it] + Training...: 68% 2987/4393 [4:05:47<1:46:46, 4.56s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:42:52<17:31:17, 21025.84s/it] + Training...: 68% 2987/4393 [4:05:52<1:46:46, 4.56s/it] + Training...: 68% 2988/4393 [4:05:52<1:48:18, 4.63s/it] + Training...: 68% 2989/4393 [4:05:56<1:43:41, 4.43s/it] + Training...: 68% 2990/4393 [4:06:00<1:38:57, 4.23s/it] + Training...: 68% 2991/4393 [4:06:04<1:34:04, 4.03s/it] + Training...: 68% 2992/4393 [4:06:07<1:28:15, 3.78s/it] + Training...: 68% 2993/4393 [4:06:10<1:21:53, 3.51s/it] + Training...: 68% 2994/4393 [4:06:12<1:15:31, 3.24s/it] + Training...: 68% 2995/4393 [4:06:15<1:09:10, 2.97s/it] + Training...: 68% 2996/4393 [4:06:17<1:02:53, 2.70s/it] + Training...: 68% 2997/4393 [4:06:18<56:32, 2.43s/it]  + Training...: 68% 2998/4393 [4:06:20<50:31, 2.17s/it] + Training...: 68% 2999/4393 [4:06:21<44:33, 1.92s/it] + Training...: 68% 3000/4393 [4:06:22<38:32, 1.66s/it] + Training...: 68% 3001/4393 [4:06:29<1:10:20, 3.03s/it] + Training...: 68% 3002/4393 [4:06:35<1:32:48, 4.00s/it] + Training...: 68% 3003/4393 [4:06:41<1:47:34, 4.64s/it] + Training...: 68% 3004/4393 [4:06:47<1:57:07, 5.06s/it] + Training...: 68% 3005/4393 [4:06:53<2:05:24, 5.42s/it] + Training...: 68% 3006/4393 [4:06:59<2:09:14, 5.59s/it] + Training...: 68% 3007/4393 [4:07:05<2:11:46, 5.70s/it] + Training...: 68% 3008/4393 [4:07:11<2:13:35, 5.79s/it] + Training...: 68% 3009/4393 [4:07:17<2:14:44, 5.84s/it] + Training...: 69% 3010/4393 [4:07:23<2:14:27, 5.83s/it] + Training...: 69% 3011/4393 [4:07:29<2:13:56, 5.81s/it] + Training...: 69% 3012/4393 [4:07:35<2:13:33, 5.80s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:44:41<17:31:17, 21025.84s/it] + Training...: 69% 3012/4393 [4:07:41<2:13:33, 5.80s/it] + Training...: 69% 3013/4393 [4:07:41<2:18:52, 6.04s/it] + Training...: 69% 3014/4393 [4:07:47<2:16:53, 5.96s/it] + Training...: 69% 3015/4393 [4:07:53<2:14:57, 5.88s/it] + Training...: 69% 3016/4393 [4:07:58<2:12:58, 5.79s/it] + Training...: 69% 3017/4393 [4:08:04<2:11:07, 5.72s/it] + Training...: 69% 3018/4393 [4:08:09<2:09:36, 5.66s/it] + Training...: 69% 3019/4393 [4:08:15<2:08:45, 5.62s/it] + Training...: 69% 3020/4393 [4:08:20<2:07:18, 5.56s/it] + Training...: 69% 3021/4393 [4:08:26<2:08:21, 5.61s/it] + Training...: 69% 3022/4393 [4:08:31<2:07:06, 5.56s/it] + Training...: 69% 3023/4393 [4:08:37<2:06:31, 5.54s/it] + Training...: 69% 3024/4393 [4:08:42<2:05:03, 5.48s/it] + Training...: 69% 3025/4393 [4:08:48<2:03:46, 5.43s/it] + Training...: 69% 3026/4393 [4:08:53<2:01:58, 5.35s/it] + Training...: 69% 3027/4393 [4:08:58<2:00:35, 5.30s/it] + Training...: 69% 3028/4393 [4:09:03<1:59:09, 5.24s/it] + Training...: 69% 3029/4393 [4:09:08<1:57:57, 5.19s/it] + Training...: 69% 3030/4393 [4:09:13<1:56:35, 5.13s/it] + Training...: 69% 3031/4393 [4:09:18<1:55:02, 5.07s/it] + Training...: 69% 3032/4393 [4:09:23<1:52:48, 4.97s/it] + Training...: 69% 3033/4393 [4:09:28<1:51:06, 4.90s/it] + Training...: 69% 3034/4393 [4:09:32<1:49:12, 4.82s/it] + Training...: 69% 3035/4393 [4:09:37<1:47:36, 4.75s/it] + Training...: 69% 3036/4393 [4:09:41<1:44:48, 4.63s/it] + Training...: 69% 3037/4393 [4:09:45<1:42:12, 4.52s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:46:50<17:31:17, 21025.84s/it] + Training...: 69% 3037/4393 [4:09:50<1:42:12, 4.52s/it] + Training...: 69% 3038/4393 [4:09:50<1:43:26, 4.58s/it] + Training...: 69% 3039/4393 [4:09:54<1:38:38, 4.37s/it] + Training...: 69% 3040/4393 [4:09:58<1:33:32, 4.15s/it] + Training...: 69% 3041/4393 [4:10:01<1:27:53, 3.90s/it] + Training...: 69% 3042/4393 [4:10:04<1:22:45, 3.68s/it] + Training...: 69% 3043/4393 [4:10:07<1:17:26, 3.44s/it] + Training...: 69% 3044/4393 [4:10:10<1:11:18, 3.17s/it] + Training...: 69% 3045/4393 [4:10:12<1:05:15, 2.90s/it] + Training...: 69% 3046/4393 [4:10:14<59:22, 2.65s/it]  + Training...: 69% 3047/4393 [4:10:16<53:28, 2.38s/it] + Training...: 69% 3048/4393 [4:10:17<47:42, 2.13s/it] + Training...: 69% 3049/4393 [4:10:18<41:50, 1.87s/it] + Training...: 69% 3050/4393 [4:10:19<35:56, 1.61s/it] + Training...: 69% 3051/4393 [4:10:26<1:07:40, 3.03s/it] + Training...: 69% 3052/4393 [4:10:32<1:29:31, 4.01s/it] + Training...: 69% 3053/4393 [4:10:38<1:43:26, 4.63s/it] + Training...: 70% 3054/4393 [4:10:44<1:52:11, 5.03s/it] + Training...: 70% 3055/4393 [4:10:50<1:58:15, 5.30s/it] + Training...: 70% 3056/4393 [4:10:56<2:02:47, 5.51s/it] + Training...: 70% 3057/4393 [4:11:02<2:06:14, 5.67s/it] + Training...: 70% 3058/4393 [4:11:08<2:08:12, 5.76s/it] + Training...: 70% 3059/4393 [4:11:14<2:09:21, 5.82s/it] + Training...: 70% 3060/4393 [4:11:20<2:09:53, 5.85s/it] + Training...: 70% 3061/4393 [4:11:26<2:09:53, 5.85s/it] + Training...: 70% 3062/4393 [4:11:32<2:09:22, 5.83s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:48:38<17:31:17, 21025.84s/it] + Training...: 70% 3062/4393 [4:11:38<2:09:22, 5.83s/it] + Training...: 70% 3063/4393 [4:11:38<2:14:31, 6.07s/it] + Training...: 70% 3064/4393 [4:11:44<2:12:35, 5.99s/it] + Training...: 70% 3065/4393 [4:11:50<2:11:47, 5.95s/it] + Training...: 70% 3066/4393 [4:11:55<2:09:40, 5.86s/it] + Training...: 70% 3067/4393 [4:12:01<2:07:41, 5.78s/it] + Training...: 70% 3068/4393 [4:12:07<2:06:04, 5.71s/it] + Training...: 70% 3069/4393 [4:12:12<2:04:34, 5.65s/it] + Training...: 70% 3070/4393 [4:12:18<2:02:59, 5.58s/it] + Training...: 70% 3071/4393 [4:12:23<2:02:06, 5.54s/it] + Training...: 70% 3072/4393 [4:12:28<2:00:48, 5.49s/it] + Training...: 70% 3073/4393 [4:12:34<1:59:54, 5.45s/it] + Training...: 70% 3074/4393 [4:12:39<1:58:45, 5.40s/it] + Training...: 70% 3075/4393 [4:12:44<1:57:50, 5.36s/it] + Training...: 70% 3076/4393 [4:12:49<1:56:10, 5.29s/it] + Training...: 70% 3077/4393 [4:12:55<1:54:48, 5.23s/it] + Training...: 70% 3078/4393 [4:13:00<1:53:30, 5.18s/it] + Training...: 70% 3079/4393 [4:13:05<1:52:24, 5.13s/it] + Training...: 70% 3080/4393 [4:13:10<1:51:00, 5.07s/it] + Training...: 70% 3081/4393 [4:13:14<1:49:55, 5.03s/it] + Training...: 70% 3082/4393 [4:13:19<1:48:16, 4.96s/it] + Training...: 70% 3083/4393 [4:13:24<1:46:59, 4.90s/it] + Training...: 70% 3084/4393 [4:13:29<1:44:52, 4.81s/it] + Training...: 70% 3085/4393 [4:13:33<1:43:19, 4.74s/it] + Training...: 70% 3086/4393 [4:13:38<1:41:38, 4.67s/it] + Training...: 70% 3087/4393 [4:13:42<1:39:19, 4.56s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:50:47<17:31:17, 21025.84s/it] + Training...: 70% 3087/4393 [4:13:47<1:39:19, 4.56s/it] + Training...: 70% 3088/4393 [4:13:47<1:41:03, 4.65s/it] + Training...: 70% 3089/4393 [4:13:51<1:36:55, 4.46s/it] + Training...: 70% 3090/4393 [4:13:55<1:32:18, 4.25s/it] + Training...: 70% 3091/4393 [4:13:58<1:27:38, 4.04s/it] + Training...: 70% 3092/4393 [4:14:02<1:23:12, 3.84s/it] + Training...: 70% 3093/4393 [4:14:05<1:18:32, 3.63s/it] + Training...: 70% 3094/4393 [4:14:08<1:13:26, 3.39s/it] + Training...: 70% 3095/4393 [4:14:10<1:07:19, 3.11s/it] + Training...: 70% 3096/4393 [4:14:12<1:01:15, 2.83s/it] + Training...: 70% 3097/4393 [4:14:14<54:43, 2.53s/it]  + Training...: 71% 3098/4393 [4:14:16<48:37, 2.25s/it] + Training...: 71% 3099/4393 [4:14:17<42:21, 1.96s/it] + Training...: 71% 3100/4393 [4:14:18<36:08, 1.68s/it] + Training...: 71% 3101/4393 [4:14:24<1:05:30, 3.04s/it] + Training...: 71% 3102/4393 [4:14:30<1:26:27, 4.02s/it] + Training...: 71% 3103/4393 [4:14:36<1:39:44, 4.64s/it] + Training...: 71% 3104/4393 [4:14:43<1:48:53, 5.07s/it] + Training...: 71% 3105/4393 [4:14:49<1:54:52, 5.35s/it] + Training...: 71% 3106/4393 [4:14:54<1:58:27, 5.52s/it] + Training...: 71% 3107/4393 [4:15:00<2:00:56, 5.64s/it] + Training...: 71% 3108/4393 [4:15:06<2:02:50, 5.74s/it] + Training...: 71% 3109/4393 [4:15:12<2:03:32, 5.77s/it] + Training...: 71% 3110/4393 [4:15:18<2:03:52, 5.79s/it] + Training...: 71% 3111/4393 [4:15:24<2:03:52, 5.80s/it] + Training...: 71% 3112/4393 [4:15:30<2:03:18, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:52:36<17:31:17, 21025.84s/it] + Training...: 71% 3112/4393 [4:15:36<2:03:18, 5.78s/it] + Training...: 71% 3113/4393 [4:15:36<2:08:04, 6.00s/it] + Training...: 71% 3114/4393 [4:15:42<2:06:19, 5.93s/it] + Training...: 71% 3115/4393 [4:15:48<2:06:32, 5.94s/it] + Training...: 71% 3116/4393 [4:15:53<2:04:29, 5.85s/it] + Training...: 71% 3117/4393 [4:15:59<2:02:23, 5.76s/it] + Training...: 71% 3118/4393 [4:16:05<2:00:43, 5.68s/it] + Training...: 71% 3119/4393 [4:16:10<1:59:32, 5.63s/it] + Training...: 71% 3120/4393 [4:16:16<1:58:24, 5.58s/it] + Training...: 71% 3121/4393 [4:16:21<1:57:06, 5.52s/it] + Training...: 71% 3122/4393 [4:16:26<1:56:11, 5.49s/it] + Training...: 71% 3123/4393 [4:16:32<1:55:16, 5.45s/it] + Training...: 71% 3124/4393 [4:16:37<1:54:08, 5.40s/it] + Training...: 71% 3125/4393 [4:16:42<1:53:20, 5.36s/it] + Training...: 71% 3126/4393 [4:16:47<1:51:56, 5.30s/it] + Training...: 71% 3127/4393 [4:16:53<1:51:00, 5.26s/it] + Training...: 71% 3128/4393 [4:16:58<1:49:56, 5.21s/it] + Training...: 71% 3129/4393 [4:17:03<1:48:39, 5.16s/it] + Training...: 71% 3130/4393 [4:17:08<1:47:05, 5.09s/it] + Training...: 71% 3131/4393 [4:17:13<1:46:21, 5.06s/it] + Training...: 71% 3132/4393 [4:17:17<1:44:24, 4.97s/it] + Training...: 71% 3133/4393 [4:17:22<1:42:51, 4.90s/it] + Training...: 71% 3134/4393 [4:17:27<1:41:10, 4.82s/it] + Training...: 71% 3135/4393 [4:17:31<1:39:06, 4.73s/it] + Training...: 71% 3136/4393 [4:17:36<1:37:03, 4.63s/it] + Training...: 71% 3137/4393 [4:17:40<1:35:17, 4.55s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:54:45<17:31:17, 21025.84s/it] + Training...: 71% 3137/4393 [4:17:45<1:35:17, 4.55s/it] + Training...: 71% 3138/4393 [4:17:45<1:36:23, 4.61s/it] + Training...: 71% 3139/4393 [4:17:49<1:31:35, 4.38s/it] + Training...: 71% 3140/4393 [4:17:52<1:26:49, 4.16s/it] + Training...: 72% 3141/4393 [4:17:56<1:22:08, 3.94s/it] + Training...: 72% 3142/4393 [4:17:59<1:16:54, 3.69s/it] + Training...: 72% 3143/4393 [4:18:02<1:11:59, 3.46s/it] + Training...: 72% 3144/4393 [4:18:04<1:06:45, 3.21s/it] + Training...: 72% 3145/4393 [4:18:07<1:01:54, 2.98s/it] + Training...: 72% 3146/4393 [4:18:09<56:35, 2.72s/it]  + Training...: 72% 3147/4393 [4:18:11<51:03, 2.46s/it] + Training...: 72% 3148/4393 [4:18:12<45:36, 2.20s/it] + Training...: 72% 3149/4393 [4:18:14<39:53, 1.92s/it] + Training...: 72% 3150/4393 [4:18:15<34:20, 1.66s/it] + Training...: 72% 3151/4393 [4:18:21<1:02:33, 3.02s/it] + Training...: 72% 3152/4393 [4:18:27<1:22:16, 3.98s/it] + Training...: 72% 3153/4393 [4:18:33<1:35:13, 4.61s/it] + Training...: 72% 3154/4393 [4:18:39<1:43:30, 5.01s/it] + Training...: 72% 3155/4393 [4:18:45<1:50:11, 5.34s/it] + Training...: 72% 3156/4393 [4:18:51<1:53:58, 5.53s/it] + Training...: 72% 3157/4393 [4:18:57<1:56:46, 5.67s/it] + Training...: 72% 3158/4393 [4:19:03<1:58:09, 5.74s/it] + Training...: 72% 3159/4393 [4:19:09<1:58:42, 5.77s/it] + Training...: 72% 3160/4393 [4:19:15<1:58:49, 5.78s/it] + Training...: 72% 3161/4393 [4:19:21<1:59:31, 5.82s/it] + Training...: 72% 3162/4393 [4:19:27<2:00:46, 5.89s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:56:33<17:31:17, 21025.84s/it] + Training...: 72% 3162/4393 [4:19:33<2:00:46, 5.89s/it] + Training...: 72% 3163/4393 [4:19:33<2:04:59, 6.10s/it] + Training...: 72% 3164/4393 [4:19:39<2:02:39, 5.99s/it] + Training...: 72% 3165/4393 [4:19:45<2:00:27, 5.89s/it] + Training...: 72% 3166/4393 [4:19:50<1:58:44, 5.81s/it] + Training...: 72% 3167/4393 [4:19:56<1:57:19, 5.74s/it] + Training...: 72% 3168/4393 [4:20:01<1:55:48, 5.67s/it] + Training...: 72% 3169/4393 [4:20:07<1:55:04, 5.64s/it] + Training...: 72% 3170/4393 [4:20:12<1:53:58, 5.59s/it] + Training...: 72% 3171/4393 [4:20:18<1:53:30, 5.57s/it] + Training...: 72% 3172/4393 [4:20:23<1:51:54, 5.50s/it] + Training...: 72% 3173/4393 [4:20:29<1:50:54, 5.45s/it] + Training...: 72% 3174/4393 [4:20:34<1:50:16, 5.43s/it] + Training...: 72% 3175/4393 [4:20:39<1:49:42, 5.40s/it] + Training...: 72% 3176/4393 [4:20:45<1:48:46, 5.36s/it] + Training...: 72% 3177/4393 [4:20:50<1:48:31, 5.35s/it] + Training...: 72% 3178/4393 [4:20:55<1:47:19, 5.30s/it] + Training...: 72% 3179/4393 [4:21:00<1:45:36, 5.22s/it] + Training...: 72% 3180/4393 [4:21:05<1:43:56, 5.14s/it] + Training...: 72% 3181/4393 [4:21:10<1:42:26, 5.07s/it] + Training...: 72% 3182/4393 [4:21:15<1:41:00, 5.00s/it] + Training...: 72% 3183/4393 [4:21:20<1:39:18, 4.92s/it] + Training...: 72% 3184/4393 [4:21:24<1:37:07, 4.82s/it] + Training...: 73% 3185/4393 [4:21:29<1:35:20, 4.74s/it] + Training...: 73% 3186/4393 [4:21:33<1:33:12, 4.63s/it] + Training...: 73% 3187/4393 [4:21:37<1:30:35, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [56:58:42<17:31:17, 21025.84s/it] + Training...: 73% 3187/4393 [4:21:42<1:30:35, 4.51s/it] + Training...: 73% 3188/4393 [4:21:42<1:31:26, 4.55s/it] + Training...: 73% 3189/4393 [4:21:46<1:27:30, 4.36s/it] + Training...: 73% 3190/4393 [4:21:50<1:23:17, 4.15s/it] + Training...: 73% 3191/4393 [4:21:53<1:18:51, 3.94s/it] + Training...: 73% 3192/4393 [4:21:56<1:13:53, 3.69s/it] + Training...: 73% 3193/4393 [4:21:59<1:09:01, 3.45s/it] + Training...: 73% 3194/4393 [4:22:02<1:03:40, 3.19s/it] + Training...: 73% 3195/4393 [4:22:04<58:25, 2.93s/it]  + Training...: 73% 3196/4393 [4:22:06<53:17, 2.67s/it] + Training...: 73% 3197/4393 [4:22:08<48:05, 2.41s/it] + Training...: 73% 3198/4393 [4:22:09<43:09, 2.17s/it] + Training...: 73% 3199/4393 [4:22:11<38:05, 1.91s/it] + Training...: 73% 3200/4393 [4:22:12<32:57, 1.66s/it] + Training...: 73% 3201/4393 [4:22:18<1:00:25, 3.04s/it] + Training...: 73% 3202/4393 [4:22:24<1:19:17, 3.99s/it] + Training...: 73% 3203/4393 [4:22:30<1:32:15, 4.65s/it] + Training...: 73% 3204/4393 [4:22:37<1:41:49, 5.14s/it] + Training...: 73% 3205/4393 [4:22:43<1:46:53, 5.40s/it] + Training...: 73% 3206/4393 [4:22:49<1:50:00, 5.56s/it] + Training...: 73% 3207/4393 [4:22:55<1:51:53, 5.66s/it] + Training...: 73% 3208/4393 [4:23:00<1:53:05, 5.73s/it] + Training...: 73% 3209/4393 [4:23:06<1:54:04, 5.78s/it] + Training...: 73% 3210/4393 [4:23:12<1:53:47, 5.77s/it] + Training...: 73% 3211/4393 [4:23:18<1:54:29, 5.81s/it] + Training...: 73% 3212/4393 [4:23:24<1:54:42, 5.83s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:00:30<17:31:17, 21025.84s/it] + Training...: 73% 3212/4393 [4:23:30<1:54:42, 5.83s/it] + Training...: 73% 3213/4393 [4:23:30<1:59:01, 6.05s/it] + Training...: 73% 3214/4393 [4:23:36<1:56:53, 5.95s/it] + Training...: 73% 3215/4393 [4:23:42<1:55:13, 5.87s/it] + Training...: 73% 3216/4393 [4:23:47<1:53:36, 5.79s/it] + Training...: 73% 3217/4393 [4:23:53<1:53:26, 5.79s/it] + Training...: 73% 3218/4393 [4:23:59<1:53:01, 5.77s/it] + Training...: 73% 3219/4393 [4:24:04<1:51:22, 5.69s/it] + Training...: 73% 3220/4393 [4:24:10<1:49:47, 5.62s/it] + Training...: 73% 3221/4393 [4:24:15<1:48:39, 5.56s/it] + Training...: 73% 3222/4393 [4:24:21<1:47:15, 5.50s/it] + Training...: 73% 3223/4393 [4:24:26<1:46:18, 5.45s/it] + Training...: 73% 3224/4393 [4:24:31<1:45:09, 5.40s/it] + Training...: 73% 3225/4393 [4:24:36<1:44:10, 5.35s/it] + Training...: 73% 3226/4393 [4:24:42<1:42:58, 5.29s/it] + Training...: 73% 3227/4393 [4:24:47<1:42:15, 5.26s/it] + Training...: 73% 3228/4393 [4:24:52<1:41:15, 5.22s/it] + Training...: 74% 3229/4393 [4:24:57<1:40:24, 5.18s/it] + Training...: 74% 3230/4393 [4:25:02<1:38:54, 5.10s/it] + Training...: 74% 3231/4393 [4:25:07<1:37:38, 5.04s/it] + Training...: 74% 3232/4393 [4:25:12<1:36:21, 4.98s/it] + Training...: 74% 3233/4393 [4:25:17<1:35:17, 4.93s/it] + Training...: 74% 3234/4393 [4:25:21<1:34:40, 4.90s/it] + Training...: 74% 3235/4393 [4:25:26<1:33:19, 4.84s/it] + Training...: 74% 3236/4393 [4:25:30<1:30:45, 4.71s/it] + Training...: 74% 3237/4393 [4:25:35<1:28:27, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:02:39<17:31:17, 21025.84s/it] + Training...: 74% 3237/4393 [4:25:39<1:28:27, 4.59s/it] + Training...: 74% 3238/4393 [4:25:39<1:28:55, 4.62s/it] + Training...: 74% 3239/4393 [4:25:43<1:24:47, 4.41s/it] + Training...: 74% 3240/4393 [4:25:47<1:20:39, 4.20s/it] + Training...: 74% 3241/4393 [4:25:51<1:16:34, 3.99s/it] + Training...: 74% 3242/4393 [4:25:54<1:11:54, 3.75s/it] + Training...: 74% 3243/4393 [4:25:57<1:07:37, 3.53s/it] + Training...: 74% 3244/4393 [4:26:00<1:03:04, 3.29s/it] + Training...: 74% 3245/4393 [4:26:02<58:15, 3.04s/it]  + Training...: 74% 3246/4393 [4:26:04<53:04, 2.78s/it] + Training...: 74% 3247/4393 [4:26:06<47:48, 2.50s/it] + Training...: 74% 3248/4393 [4:26:08<42:39, 2.24s/it] + Training...: 74% 3249/4393 [4:26:09<37:11, 1.95s/it] + Training...: 74% 3250/4393 [4:26:10<31:50, 1.67s/it] + Training...: 74% 3251/4393 [4:26:16<57:29, 3.02s/it] + Training...: 74% 3252/4393 [4:26:22<1:15:58, 4.00s/it] + Training...: 74% 3253/4393 [4:26:28<1:28:09, 4.64s/it] + Training...: 74% 3254/4393 [4:26:34<1:35:49, 5.05s/it] + Training...: 74% 3255/4393 [4:26:41<1:41:28, 5.35s/it] + Training...: 74% 3256/4393 [4:26:46<1:44:45, 5.53s/it] + Training...: 74% 3257/4393 [4:26:52<1:46:50, 5.64s/it] + Training...: 74% 3258/4393 [4:26:58<1:48:11, 5.72s/it] + Training...: 74% 3259/4393 [4:27:04<1:49:02, 5.77s/it] + Training...: 74% 3260/4393 [4:27:10<1:49:06, 5.78s/it] + Training...: 74% 3261/4393 [4:27:16<1:49:38, 5.81s/it] + Training...: 74% 3262/4393 [4:27:22<1:49:07, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:04:28<17:31:17, 21025.84s/it] + Training...: 74% 3262/4393 [4:27:28<1:49:07, 5.79s/it] + Training...: 74% 3263/4393 [4:27:28<1:54:02, 6.06s/it] + Training...: 74% 3264/4393 [4:27:34<1:53:29, 6.03s/it] + Training...: 74% 3265/4393 [4:27:40<1:51:51, 5.95s/it] + Training...: 74% 3266/4393 [4:27:46<1:50:14, 5.87s/it] + Training...: 74% 3267/4393 [4:27:51<1:48:33, 5.78s/it] + Training...: 74% 3268/4393 [4:27:57<1:47:18, 5.72s/it] + Training...: 74% 3269/4393 [4:28:02<1:45:52, 5.65s/it] + Training...: 74% 3270/4393 [4:28:08<1:44:38, 5.59s/it] + Training...: 74% 3271/4393 [4:28:13<1:43:27, 5.53s/it] + Training...: 74% 3272/4393 [4:28:19<1:42:15, 5.47s/it] + Training...: 75% 3273/4393 [4:28:24<1:41:29, 5.44s/it] + Training...: 75% 3274/4393 [4:28:29<1:40:34, 5.39s/it] + Training...: 75% 3275/4393 [4:28:34<1:39:49, 5.36s/it] + Training...: 75% 3276/4393 [4:28:40<1:38:56, 5.31s/it] + Training...: 75% 3277/4393 [4:28:45<1:38:06, 5.27s/it] + Training...: 75% 3278/4393 [4:28:50<1:36:59, 5.22s/it] + Training...: 75% 3279/4393 [4:28:55<1:36:06, 5.18s/it] + Training...: 75% 3280/4393 [4:29:00<1:34:34, 5.10s/it] + Training...: 75% 3281/4393 [4:29:05<1:33:58, 5.07s/it] + Training...: 75% 3282/4393 [4:29:10<1:33:12, 5.03s/it] + Training...: 75% 3283/4393 [4:29:15<1:31:54, 4.97s/it] + Training...: 75% 3284/4393 [4:29:19<1:29:58, 4.87s/it] + Training...: 75% 3285/4393 [4:29:24<1:28:08, 4.77s/it] + Training...: 75% 3286/4393 [4:29:28<1:25:51, 4.65s/it] + Training...: 75% 3287/4393 [4:29:32<1:23:24, 4.53s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:06:37<17:31:17, 21025.84s/it] + Training...: 75% 3287/4393 [4:29:37<1:23:24, 4.53s/it] + Training...: 75% 3288/4393 [4:29:37<1:24:09, 4.57s/it] + Training...: 75% 3289/4393 [4:29:41<1:19:55, 4.34s/it] + Training...: 75% 3290/4393 [4:29:45<1:15:32, 4.11s/it] + Training...: 75% 3291/4393 [4:29:48<1:11:30, 3.89s/it] + Training...: 75% 3292/4393 [4:29:51<1:06:43, 3.64s/it] + Training...: 75% 3293/4393 [4:29:54<1:02:04, 3.39s/it] + Training...: 75% 3294/4393 [4:29:56<57:32, 3.14s/it]  + Training...: 75% 3295/4393 [4:29:59<52:57, 2.89s/it] + Training...: 75% 3296/4393 [4:30:01<48:21, 2.64s/it] + Training...: 75% 3297/4393 [4:30:03<43:41, 2.39s/it] + Training...: 75% 3298/4393 [4:30:04<38:57, 2.14s/it] + Training...: 75% 3299/4393 [4:30:05<34:02, 1.87s/it] + Training...: 75% 3300/4393 [4:30:06<29:08, 1.60s/it] + Training...: 75% 3301/4393 [4:30:13<54:46, 3.01s/it] + Training...: 75% 3302/4393 [4:30:19<1:12:23, 3.98s/it] + Training...: 75% 3303/4393 [4:30:25<1:24:00, 4.62s/it] + Training...: 75% 3304/4393 [4:30:31<1:31:27, 5.04s/it] + Training...: 75% 3305/4393 [4:30:37<1:37:46, 5.39s/it] + Training...: 75% 3306/4393 [4:30:43<1:41:38, 5.61s/it] + Training...: 75% 3307/4393 [4:30:49<1:43:30, 5.72s/it] + Training...: 75% 3308/4393 [4:30:55<1:44:21, 5.77s/it] + Training...: 75% 3309/4393 [4:31:01<1:45:01, 5.81s/it] + Training...: 75% 3310/4393 [4:31:07<1:44:54, 5.81s/it] + Training...: 75% 3311/4393 [4:31:13<1:44:33, 5.80s/it] + Training...: 75% 3312/4393 [4:31:18<1:44:12, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:08:25<17:31:17, 21025.84s/it] + Training...: 75% 3312/4393 [4:31:25<1:44:12, 5.78s/it] + Training...: 75% 3313/4393 [4:31:25<1:48:27, 6.03s/it] + Training...: 75% 3314/4393 [4:31:31<1:46:53, 5.94s/it] + Training...: 75% 3315/4393 [4:31:36<1:45:39, 5.88s/it] + Training...: 75% 3316/4393 [4:31:42<1:44:37, 5.83s/it] + Training...: 76% 3317/4393 [4:31:48<1:43:18, 5.76s/it] + Training...: 76% 3318/4393 [4:31:53<1:42:12, 5.71s/it] + Training...: 76% 3319/4393 [4:31:59<1:41:30, 5.67s/it] + Training...: 76% 3320/4393 [4:32:04<1:40:24, 5.61s/it] + Training...: 76% 3321/4393 [4:32:10<1:39:46, 5.58s/it] + Training...: 76% 3322/4393 [4:32:16<1:40:00, 5.60s/it] + Training...: 76% 3323/4393 [4:32:21<1:39:00, 5.55s/it] + Training...: 76% 3324/4393 [4:32:26<1:37:27, 5.47s/it] + Training...: 76% 3325/4393 [4:32:32<1:36:24, 5.42s/it] + Training...: 76% 3326/4393 [4:32:37<1:35:32, 5.37s/it] + Training...: 76% 3327/4393 [4:32:42<1:34:48, 5.34s/it] + Training...: 76% 3328/4393 [4:32:47<1:33:21, 5.26s/it] + Training...: 76% 3329/4393 [4:32:52<1:32:05, 5.19s/it] + Training...: 76% 3330/4393 [4:32:57<1:31:14, 5.15s/it] + Training...: 76% 3331/4393 [4:33:02<1:29:50, 5.08s/it] + Training...: 76% 3332/4393 [4:33:07<1:28:14, 4.99s/it] + Training...: 76% 3333/4393 [4:33:12<1:27:06, 4.93s/it] + Training...: 76% 3334/4393 [4:33:16<1:25:32, 4.85s/it] + Training...: 76% 3335/4393 [4:33:21<1:24:06, 4.77s/it] + Training...: 76% 3336/4393 [4:33:25<1:21:55, 4.65s/it] + Training...: 76% 3337/4393 [4:33:30<1:19:28, 4.52s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:10:34<17:31:17, 21025.84s/it] + Training...: 76% 3337/4393 [4:33:34<1:19:28, 4.52s/it] + Training...: 76% 3338/4393 [4:33:34<1:19:58, 4.55s/it] + Training...: 76% 3339/4393 [4:33:38<1:16:18, 4.34s/it] + Training...: 76% 3340/4393 [4:33:42<1:12:02, 4.11s/it] + Training...: 76% 3341/4393 [4:33:45<1:07:57, 3.88s/it] + Training...: 76% 3342/4393 [4:33:48<1:03:21, 3.62s/it] + Training...: 76% 3343/4393 [4:33:51<58:47, 3.36s/it]  + Training...: 76% 3344/4393 [4:33:53<54:10, 3.10s/it] + Training...: 76% 3345/4393 [4:33:56<49:58, 2.86s/it] + Training...: 76% 3346/4393 [4:33:58<45:42, 2.62s/it] + Training...: 76% 3347/4393 [4:33:59<41:13, 2.37s/it] + Training...: 76% 3348/4393 [4:34:01<36:56, 2.12s/it] + Training...: 76% 3349/4393 [4:34:02<32:48, 1.89s/it] + Training...: 76% 3350/4393 [4:34:03<28:24, 1.63s/it] + Training...: 76% 3351/4393 [4:34:10<53:13, 3.06s/it] + Training...: 76% 3352/4393 [4:34:16<1:10:18, 4.05s/it] + Training...: 76% 3353/4393 [4:34:22<1:20:45, 4.66s/it] + Training...: 76% 3354/4393 [4:34:28<1:28:05, 5.09s/it] + Training...: 76% 3355/4393 [4:34:34<1:33:15, 5.39s/it] + Training...: 76% 3356/4393 [4:34:40<1:35:56, 5.55s/it] + Training...: 76% 3357/4393 [4:34:46<1:37:43, 5.66s/it] + Training...: 76% 3358/4393 [4:34:52<1:38:38, 5.72s/it] + Training...: 76% 3359/4393 [4:34:58<1:39:18, 5.76s/it] + Training...: 76% 3360/4393 [4:35:04<1:39:31, 5.78s/it] + Training...: 77% 3361/4393 [4:35:09<1:39:34, 5.79s/it] + Training...: 77% 3362/4393 [4:35:15<1:39:11, 5.77s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:12:22<17:31:17, 21025.84s/it] + Training...: 77% 3362/4393 [4:35:22<1:39:11, 5.77s/it] + Training...: 77% 3363/4393 [4:35:22<1:43:19, 6.02s/it] + Training...: 77% 3364/4393 [4:35:28<1:42:04, 5.95s/it] + Training...: 77% 3365/4393 [4:35:33<1:40:54, 5.89s/it] + Training...: 77% 3366/4393 [4:35:39<1:39:24, 5.81s/it] + Training...: 77% 3367/4393 [4:35:45<1:38:18, 5.75s/it] + Training...: 77% 3368/4393 [4:35:50<1:37:52, 5.73s/it] + Training...: 77% 3369/4393 [4:35:56<1:37:19, 5.70s/it] + Training...: 77% 3370/4393 [4:36:01<1:36:08, 5.64s/it] + Training...: 77% 3371/4393 [4:36:07<1:34:50, 5.57s/it] + Training...: 77% 3372/4393 [4:36:12<1:33:25, 5.49s/it] + Training...: 77% 3373/4393 [4:36:17<1:32:34, 5.45s/it] + Training...: 77% 3374/4393 [4:36:23<1:31:39, 5.40s/it] + Training...: 77% 3375/4393 [4:36:28<1:31:00, 5.36s/it] + Training...: 77% 3376/4393 [4:36:33<1:29:54, 5.30s/it] + Training...: 77% 3377/4393 [4:36:38<1:29:31, 5.29s/it] + Training...: 77% 3378/4393 [4:36:43<1:28:16, 5.22s/it] + Training...: 77% 3379/4393 [4:36:49<1:27:32, 5.18s/it] + Training...: 77% 3380/4393 [4:36:54<1:26:15, 5.11s/it] + Training...: 77% 3381/4393 [4:36:58<1:25:18, 5.06s/it] + Training...: 77% 3382/4393 [4:37:03<1:24:19, 5.00s/it] + Training...: 77% 3383/4393 [4:37:08<1:23:02, 4.93s/it] + Training...: 77% 3384/4393 [4:37:13<1:21:30, 4.85s/it] + Training...: 77% 3385/4393 [4:37:17<1:20:06, 4.77s/it] + Training...: 77% 3386/4393 [4:37:22<1:18:11, 4.66s/it] + Training...: 77% 3387/4393 [4:37:26<1:16:24, 4.56s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:14:31<17:31:17, 21025.84s/it] + Training...: 77% 3387/4393 [4:37:31<1:16:24, 4.56s/it] + Training...: 77% 3388/4393 [4:37:31<1:17:21, 4.62s/it] + Training...: 77% 3389/4393 [4:37:35<1:14:12, 4.44s/it] + Training...: 77% 3390/4393 [4:37:39<1:11:01, 4.25s/it] + Training...: 77% 3391/4393 [4:37:42<1:07:56, 4.07s/it] + Training...: 77% 3392/4393 [4:37:46<1:04:00, 3.84s/it] + Training...: 77% 3393/4393 [4:37:49<59:58, 3.60s/it]  + Training...: 77% 3394/4393 [4:37:51<55:48, 3.35s/it] + Training...: 77% 3395/4393 [4:37:54<51:28, 3.10s/it] + Training...: 77% 3396/4393 [4:37:56<47:09, 2.84s/it] + Training...: 77% 3397/4393 [4:37:58<42:23, 2.55s/it] + Training...: 77% 3398/4393 [4:38:00<37:54, 2.29s/it] + Training...: 77% 3399/4393 [4:38:01<33:14, 2.01s/it] + Training...: 77% 3400/4393 [4:38:02<28:31, 1.72s/it] + Training...: 77% 3401/4393 [4:38:09<51:40, 3.13s/it] + Training...: 77% 3402/4393 [4:38:15<1:07:23, 4.08s/it] + Training...: 77% 3403/4393 [4:38:21<1:17:27, 4.69s/it] + Training...: 77% 3404/4393 [4:38:27<1:23:54, 5.09s/it] + Training...: 78% 3405/4393 [4:38:33<1:28:20, 5.37s/it] + Training...: 78% 3406/4393 [4:38:39<1:31:29, 5.56s/it] + Training...: 78% 3407/4393 [4:38:45<1:33:38, 5.70s/it] + Training...: 78% 3408/4393 [4:38:51<1:34:33, 5.76s/it] + Training...: 78% 3409/4393 [4:38:57<1:35:10, 5.80s/it] + Training...: 78% 3410/4393 [4:39:03<1:34:56, 5.80s/it] + Training...: 78% 3411/4393 [4:39:08<1:35:20, 5.83s/it] + Training...: 78% 3412/4393 [4:39:14<1:35:41, 5.85s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:16:21<17:31:17, 21025.84s/it] + Training...: 78% 3412/4393 [4:39:21<1:35:41, 5.85s/it] + Training...: 78% 3413/4393 [4:39:21<1:39:59, 6.12s/it] + Training...: 78% 3414/4393 [4:39:27<1:38:14, 6.02s/it] + Training...: 78% 3415/4393 [4:39:33<1:37:02, 5.95s/it] + Training...: 78% 3416/4393 [4:39:38<1:35:38, 5.87s/it] + Training...: 78% 3417/4393 [4:39:44<1:34:15, 5.79s/it] + Training...: 78% 3418/4393 [4:39:50<1:32:36, 5.70s/it] + Training...: 78% 3419/4393 [4:39:55<1:31:28, 5.63s/it] + Training...: 78% 3420/4393 [4:40:00<1:30:13, 5.56s/it] + Training...: 78% 3421/4393 [4:40:06<1:29:26, 5.52s/it] + Training...: 78% 3422/4393 [4:40:11<1:28:31, 5.47s/it] + Training...: 78% 3423/4393 [4:40:17<1:28:00, 5.44s/it] + Training...: 78% 3424/4393 [4:40:22<1:27:18, 5.41s/it] + Training...: 78% 3425/4393 [4:40:27<1:26:38, 5.37s/it] + Training...: 78% 3426/4393 [4:40:32<1:25:50, 5.33s/it] + Training...: 78% 3427/4393 [4:40:38<1:25:08, 5.29s/it] + Training...: 78% 3428/4393 [4:40:43<1:24:19, 5.24s/it] + Training...: 78% 3429/4393 [4:40:48<1:23:51, 5.22s/it] + Training...: 78% 3430/4393 [4:40:53<1:22:50, 5.16s/it] + Training...: 78% 3431/4393 [4:40:58<1:21:51, 5.11s/it] + Training...: 78% 3432/4393 [4:41:03<1:21:07, 5.07s/it] + Training...: 78% 3433/4393 [4:41:08<1:20:35, 5.04s/it] + Training...: 78% 3434/4393 [4:41:12<1:18:43, 4.93s/it] + Training...: 78% 3435/4393 [4:41:17<1:16:58, 4.82s/it] + Training...: 78% 3436/4393 [4:41:21<1:15:01, 4.70s/it] + Training...: 78% 3437/4393 [4:41:26<1:13:10, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:18:31<17:31:17, 21025.84s/it] + Training...: 78% 3437/4393 [4:41:31<1:13:10, 4.59s/it] + Training...: 78% 3438/4393 [4:41:31<1:14:08, 4.66s/it] + Training...: 78% 3439/4393 [4:41:35<1:10:45, 4.45s/it] + Training...: 78% 3440/4393 [4:41:38<1:07:28, 4.25s/it] + Training...: 78% 3441/4393 [4:41:42<1:04:06, 4.04s/it] + Training...: 78% 3442/4393 [4:41:45<1:00:16, 3.80s/it] + Training...: 78% 3443/4393 [4:41:48<56:15, 3.55s/it]  + Training...: 78% 3444/4393 [4:41:51<52:18, 3.31s/it] + Training...: 78% 3445/4393 [4:41:53<48:08, 3.05s/it] + Training...: 78% 3446/4393 [4:41:56<43:59, 2.79s/it] + Training...: 78% 3447/4393 [4:41:57<39:26, 2.50s/it] + Training...: 78% 3448/4393 [4:41:59<35:01, 2.22s/it] + Training...: 79% 3449/4393 [4:42:00<30:44, 1.95s/it] + Training...: 79% 3450/4393 [4:42:01<26:27, 1.68s/it] + Training...: 79% 3451/4393 [4:42:08<48:14, 3.07s/it] + Training...: 79% 3452/4393 [4:42:14<1:03:31, 4.05s/it] + Training...: 79% 3453/4393 [4:42:20<1:13:15, 4.68s/it] + Training...: 79% 3454/4393 [4:42:26<1:19:43, 5.09s/it] + Training...: 79% 3455/4393 [4:42:32<1:23:54, 5.37s/it] + Training...: 79% 3456/4393 [4:42:38<1:26:37, 5.55s/it] + Training...: 79% 3457/4393 [4:42:44<1:28:34, 5.68s/it] + Training...: 79% 3458/4393 [4:42:50<1:30:43, 5.82s/it] + Training...: 79% 3459/4393 [4:42:56<1:31:12, 5.86s/it] + Training...: 79% 3460/4393 [4:43:02<1:30:48, 5.84s/it] + Training...: 79% 3461/4393 [4:43:08<1:30:26, 5.82s/it] + Training...: 79% 3462/4393 [4:43:13<1:29:45, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:20:20<17:31:17, 21025.84s/it] + Training...: 79% 3462/4393 [4:43:20<1:29:45, 5.78s/it] + Training...: 79% 3463/4393 [4:43:20<1:33:17, 6.02s/it] + Training...: 79% 3464/4393 [4:43:26<1:31:40, 5.92s/it] + Training...: 79% 3465/4393 [4:43:31<1:30:46, 5.87s/it] + Training...: 79% 3466/4393 [4:43:37<1:29:40, 5.80s/it] + Training...: 79% 3467/4393 [4:43:43<1:28:47, 5.75s/it] + Training...: 79% 3468/4393 [4:43:48<1:27:49, 5.70s/it] + Training...: 79% 3469/4393 [4:43:54<1:26:54, 5.64s/it] + Training...: 79% 3470/4393 [4:43:59<1:25:50, 5.58s/it] + Training...: 79% 3471/4393 [4:44:05<1:25:26, 5.56s/it] + Training...: 79% 3472/4393 [4:44:10<1:24:44, 5.52s/it] + Training...: 79% 3473/4393 [4:44:16<1:24:10, 5.49s/it] + Training...: 79% 3474/4393 [4:44:21<1:23:50, 5.47s/it] + Training...: 79% 3475/4393 [4:44:27<1:23:40, 5.47s/it] + Training...: 79% 3476/4393 [4:44:32<1:22:23, 5.39s/it] + Training...: 79% 3477/4393 [4:44:37<1:21:19, 5.33s/it] + Training...: 79% 3478/4393 [4:44:42<1:20:13, 5.26s/it] + Training...: 79% 3479/4393 [4:44:47<1:19:35, 5.22s/it] + Training...: 79% 3480/4393 [4:44:52<1:18:41, 5.17s/it] + Training...: 79% 3481/4393 [4:44:57<1:17:29, 5.10s/it] + Training...: 79% 3482/4393 [4:45:02<1:16:11, 5.02s/it] + Training...: 79% 3483/4393 [4:45:07<1:15:02, 4.95s/it] + Training...: 79% 3484/4393 [4:45:11<1:13:36, 4.86s/it] + Training...: 79% 3485/4393 [4:45:16<1:12:27, 4.79s/it] + Training...: 79% 3486/4393 [4:45:20<1:10:44, 4.68s/it] + Training...: 79% 3487/4393 [4:45:25<1:08:57, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:22:29<17:31:17, 21025.84s/it] + Training...: 79% 3487/4393 [4:45:30<1:08:57, 4.57s/it] + Training...: 79% 3488/4393 [4:45:30<1:09:37, 4.62s/it] + Training...: 79% 3489/4393 [4:45:33<1:06:24, 4.41s/it] + Training...: 79% 3490/4393 [4:45:37<1:02:56, 4.18s/it] + Training...: 79% 3491/4393 [4:45:40<59:24, 3.95s/it]  + Training...: 79% 3492/4393 [4:45:44<55:46, 3.71s/it] + Training...: 80% 3493/4393 [4:45:47<52:11, 3.48s/it] + Training...: 80% 3494/4393 [4:45:49<48:24, 3.23s/it] + Training...: 80% 3495/4393 [4:45:52<44:33, 2.98s/it] + Training...: 80% 3496/4393 [4:45:54<40:40, 2.72s/it] + Training...: 80% 3497/4393 [4:45:56<36:49, 2.47s/it] + Training...: 80% 3498/4393 [4:45:57<32:55, 2.21s/it] + Training...: 80% 3499/4393 [4:45:59<28:55, 1.94s/it] + Training...: 80% 3500/4393 [4:46:00<24:45, 1.66s/it] + Training...: 80% 3501/4393 [4:46:06<44:59, 3.03s/it] + Training...: 80% 3502/4393 [4:46:12<59:09, 3.98s/it] + Training...: 80% 3503/4393 [4:46:18<1:08:34, 4.62s/it] + Training...: 80% 3504/4393 [4:46:24<1:14:41, 5.04s/it] + Training...: 80% 3505/4393 [4:46:30<1:18:47, 5.32s/it] + Training...: 80% 3506/4393 [4:46:36<1:21:31, 5.51s/it] + Training...: 80% 3507/4393 [4:46:42<1:23:29, 5.65s/it] + Training...: 80% 3508/4393 [4:46:48<1:24:29, 5.73s/it] + Training...: 80% 3509/4393 [4:46:54<1:25:32, 5.81s/it] + Training...: 80% 3510/4393 [4:47:00<1:26:22, 5.87s/it] + Training...: 80% 3511/4393 [4:47:06<1:26:19, 5.87s/it] + Training...: 80% 3512/4393 [4:47:12<1:25:44, 5.84s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:24:18<17:31:17, 21025.84s/it] + Training...: 80% 3512/4393 [4:47:18<1:25:44, 5.84s/it] + Training...: 80% 3513/4393 [4:47:18<1:28:51, 6.06s/it] + Training...: 80% 3514/4393 [4:47:24<1:27:29, 5.97s/it] + Training...: 80% 3515/4393 [4:47:30<1:26:31, 5.91s/it] + Training...: 80% 3516/4393 [4:47:35<1:25:22, 5.84s/it] + Training...: 80% 3517/4393 [4:47:41<1:24:25, 5.78s/it] + Training...: 80% 3518/4393 [4:47:47<1:23:09, 5.70s/it] + Training...: 80% 3519/4393 [4:47:52<1:22:34, 5.67s/it] + Training...: 80% 3520/4393 [4:47:58<1:21:45, 5.62s/it] + Training...: 80% 3521/4393 [4:48:03<1:20:55, 5.57s/it] + Training...: 80% 3522/4393 [4:48:08<1:20:05, 5.52s/it] + Training...: 80% 3523/4393 [4:48:14<1:19:48, 5.50s/it] + Training...: 80% 3524/4393 [4:48:19<1:19:13, 5.47s/it] + Training...: 80% 3525/4393 [4:48:25<1:18:23, 5.42s/it] + Training...: 80% 3526/4393 [4:48:30<1:18:06, 5.41s/it] + Training...: 80% 3527/4393 [4:48:35<1:17:47, 5.39s/it] + Training...: 80% 3528/4393 [4:48:41<1:17:00, 5.34s/it] + Training...: 80% 3529/4393 [4:48:46<1:16:08, 5.29s/it] + Training...: 80% 3530/4393 [4:48:51<1:14:54, 5.21s/it] + Training...: 80% 3531/4393 [4:48:56<1:13:49, 5.14s/it] + Training...: 80% 3532/4393 [4:49:01<1:12:49, 5.07s/it] + Training...: 80% 3533/4393 [4:49:06<1:11:54, 5.02s/it] + Training...: 80% 3534/4393 [4:49:10<1:10:20, 4.91s/it] + Training...: 80% 3535/4393 [4:49:15<1:09:04, 4.83s/it] + Training...: 80% 3536/4393 [4:49:19<1:07:31, 4.73s/it] + Training...: 81% 3537/4393 [4:49:24<1:05:39, 4.60s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:26:28<17:31:17, 21025.84s/it] + Training...: 81% 3537/4393 [4:49:28<1:05:39, 4.60s/it] + Training...: 81% 3538/4393 [4:49:28<1:06:05, 4.64s/it] + Training...: 81% 3539/4393 [4:49:32<1:02:54, 4.42s/it] + Training...: 81% 3540/4393 [4:49:36<59:40, 4.20s/it]  + Training...: 81% 3541/4393 [4:49:39<56:22, 3.97s/it] + Training...: 81% 3542/4393 [4:49:43<52:57, 3.73s/it] + Training...: 81% 3543/4393 [4:49:45<49:20, 3.48s/it] + Training...: 81% 3544/4393 [4:49:48<45:46, 3.24s/it] + Training...: 81% 3545/4393 [4:49:50<41:54, 2.97s/it] + Training...: 81% 3546/4393 [4:49:53<38:03, 2.70s/it] + Training...: 81% 3547/4393 [4:49:54<34:06, 2.42s/it] + Training...: 81% 3548/4393 [4:49:56<30:27, 2.16s/it] + Training...: 81% 3549/4393 [4:49:57<26:56, 1.91s/it] + Training...: 81% 3550/4393 [4:49:58<23:09, 1.65s/it] + Training...: 81% 3551/4393 [4:50:05<42:29, 3.03s/it] + Training...: 81% 3552/4393 [4:50:11<56:10, 4.01s/it] + Training...: 81% 3553/4393 [4:50:17<1:05:00, 4.64s/it] + Training...: 81% 3554/4393 [4:50:23<1:10:55, 5.07s/it] + Training...: 81% 3555/4393 [4:50:29<1:15:05, 5.38s/it] + Training...: 81% 3556/4393 [4:50:35<1:18:12, 5.61s/it] + Training...: 81% 3557/4393 [4:50:41<1:19:42, 5.72s/it] + Training...: 81% 3558/4393 [4:50:47<1:20:53, 5.81s/it] + Training...: 81% 3559/4393 [4:50:53<1:21:12, 5.84s/it] + Training...: 81% 3560/4393 [4:50:59<1:20:50, 5.82s/it] + Training...: 81% 3561/4393 [4:51:05<1:20:46, 5.83s/it] + Training...: 81% 3562/4393 [4:51:11<1:20:28, 5.81s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:28:17<17:31:17, 21025.84s/it] + Training...: 81% 3562/4393 [4:51:17<1:20:28, 5.81s/it] + Training...: 81% 3563/4393 [4:51:17<1:23:29, 6.04s/it] + Training...: 81% 3564/4393 [4:51:23<1:22:09, 5.95s/it] + Training...: 81% 3565/4393 [4:51:29<1:21:12, 5.88s/it] + Training...: 81% 3566/4393 [4:51:34<1:20:19, 5.83s/it] + Training...: 81% 3567/4393 [4:51:40<1:19:21, 5.76s/it] + Training...: 81% 3568/4393 [4:51:45<1:18:22, 5.70s/it] + Training...: 81% 3569/4393 [4:51:51<1:17:32, 5.65s/it] + Training...: 81% 3570/4393 [4:51:56<1:16:44, 5.60s/it] + Training...: 81% 3571/4393 [4:52:02<1:16:10, 5.56s/it] + Training...: 81% 3572/4393 [4:52:07<1:15:21, 5.51s/it] + Training...: 81% 3573/4393 [4:52:13<1:14:48, 5.47s/it] + Training...: 81% 3574/4393 [4:52:18<1:14:48, 5.48s/it] + Training...: 81% 3575/4393 [4:52:24<1:14:29, 5.46s/it] + Training...: 81% 3576/4393 [4:52:29<1:13:11, 5.38s/it] + Training...: 81% 3577/4393 [4:52:34<1:12:18, 5.32s/it] + Training...: 81% 3578/4393 [4:52:39<1:11:21, 5.25s/it] + Training...: 81% 3579/4393 [4:52:44<1:10:23, 5.19s/it] + Training...: 81% 3580/4393 [4:52:49<1:09:37, 5.14s/it] + Training...: 82% 3581/4393 [4:52:54<1:08:51, 5.09s/it] + Training...: 82% 3582/4393 [4:52:59<1:07:28, 4.99s/it] + Training...: 82% 3583/4393 [4:53:04<1:06:16, 4.91s/it] + Training...: 82% 3584/4393 [4:53:08<1:05:03, 4.83s/it] + Training...: 82% 3585/4393 [4:53:13<1:03:42, 4.73s/it] + Training...: 82% 3586/4393 [4:53:17<1:02:24, 4.64s/it] + Training...: 82% 3587/4393 [4:53:21<1:00:59, 4.54s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:30:26<17:31:17, 21025.84s/it] + Training...: 82% 3587/4393 [4:53:26<1:00:59, 4.54s/it] + Training...: 82% 3588/4393 [4:53:26<1:01:51, 4.61s/it] + Training...: 82% 3589/4393 [4:53:30<59:08, 4.41s/it]  + Training...: 82% 3590/4393 [4:53:34<56:16, 4.20s/it] + Training...: 82% 3591/4393 [4:53:37<53:11, 3.98s/it] + Training...: 82% 3592/4393 [4:53:41<49:54, 3.74s/it] + Training...: 82% 3593/4393 [4:53:43<46:28, 3.49s/it] + Training...: 82% 3594/4393 [4:53:46<43:09, 3.24s/it] + Training...: 82% 3595/4393 [4:53:49<39:53, 3.00s/it] + Training...: 82% 3596/4393 [4:53:51<36:33, 2.75s/it] + Training...: 82% 3597/4393 [4:53:53<33:22, 2.52s/it] + Training...: 82% 3598/4393 [4:53:54<30:01, 2.27s/it] + Training...: 82% 3599/4393 [4:53:56<26:27, 2.00s/it] + Training...: 82% 3600/4393 [4:53:57<22:36, 1.71s/it] + Training...: 82% 3601/4393 [4:54:03<41:22, 3.13s/it] + Training...: 82% 3602/4393 [4:54:10<53:52, 4.09s/it] + Training...: 82% 3603/4393 [4:54:16<1:02:01, 4.71s/it] + Training...: 82% 3604/4393 [4:54:22<1:07:28, 5.13s/it] + Training...: 82% 3605/4393 [4:54:28<1:10:42, 5.38s/it] + Training...: 82% 3606/4393 [4:54:34<1:12:53, 5.56s/it] + Training...: 82% 3607/4393 [4:54:40<1:14:14, 5.67s/it] + Training...: 82% 3608/4393 [4:54:46<1:14:54, 5.73s/it] + Training...: 82% 3609/4393 [4:54:51<1:15:31, 5.78s/it] + Training...: 82% 3610/4393 [4:54:57<1:15:28, 5.78s/it] + Training...: 82% 3611/4393 [4:55:03<1:15:41, 5.81s/it] + Training...: 82% 3612/4393 [4:55:09<1:15:25, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:32:15<17:31:17, 21025.84s/it] + Training...: 82% 3612/4393 [4:55:15<1:15:25, 5.79s/it] + Training...: 82% 3613/4393 [4:55:15<1:18:14, 6.02s/it] + Training...: 82% 3614/4393 [4:55:21<1:16:57, 5.93s/it] + Training...: 82% 3615/4393 [4:55:27<1:15:59, 5.86s/it] + Training...: 82% 3616/4393 [4:55:32<1:14:58, 5.79s/it] + Training...: 82% 3617/4393 [4:55:38<1:14:40, 5.77s/it] + Training...: 82% 3618/4393 [4:55:44<1:14:22, 5.76s/it] + Training...: 82% 3619/4393 [4:55:49<1:13:26, 5.69s/it] + Training...: 82% 3620/4393 [4:55:55<1:12:18, 5.61s/it] + Training...: 82% 3621/4393 [4:56:00<1:11:22, 5.55s/it] + Training...: 82% 3622/4393 [4:56:06<1:10:37, 5.50s/it] + Training...: 82% 3623/4393 [4:56:11<1:09:54, 5.45s/it] + Training...: 82% 3624/4393 [4:56:16<1:09:18, 5.41s/it] + Training...: 83% 3625/4393 [4:56:22<1:08:43, 5.37s/it] + Training...: 83% 3626/4393 [4:56:27<1:07:57, 5.32s/it] + Training...: 83% 3627/4393 [4:56:32<1:07:30, 5.29s/it] + Training...: 83% 3628/4393 [4:56:37<1:07:01, 5.26s/it] + Training...: 83% 3629/4393 [4:56:42<1:06:25, 5.22s/it] + Training...: 83% 3630/4393 [4:56:47<1:05:29, 5.15s/it] + Training...: 83% 3631/4393 [4:56:52<1:04:56, 5.11s/it] + Training...: 83% 3632/4393 [4:56:57<1:03:55, 5.04s/it] + Training...: 83% 3633/4393 [4:57:02<1:02:59, 4.97s/it] + Training...: 83% 3634/4393 [4:57:07<1:01:56, 4.90s/it] + Training...: 83% 3635/4393 [4:57:11<1:00:47, 4.81s/it] + Training...: 83% 3636/4393 [4:57:16<59:24, 4.71s/it]  + Training...: 83% 3637/4393 [4:57:20<58:11, 4.62s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:34:25<17:31:17, 21025.84s/it] + Training...: 83% 3637/4393 [4:57:25<58:11, 4.62s/it] + Training...: 83% 3638/4393 [4:57:25<58:48, 4.67s/it] + Training...: 83% 3639/4393 [4:57:29<56:06, 4.46s/it] + Training...: 83% 3640/4393 [4:57:33<53:00, 4.22s/it] + Training...: 83% 3641/4393 [4:57:36<49:54, 3.98s/it] + Training...: 83% 3642/4393 [4:57:39<46:36, 3.72s/it] + Training...: 83% 3643/4393 [4:57:42<43:26, 3.48s/it] + Training...: 83% 3644/4393 [4:57:45<40:04, 3.21s/it] + Training...: 83% 3645/4393 [4:57:47<36:50, 2.95s/it] + Training...: 83% 3646/4393 [4:57:49<33:37, 2.70s/it] + Training...: 83% 3647/4393 [4:57:51<30:20, 2.44s/it] + Training...: 83% 3648/4393 [4:57:53<27:04, 2.18s/it] + Training...: 83% 3649/4393 [4:57:54<23:43, 1.91s/it] + Training...: 83% 3650/4393 [4:57:55<20:24, 1.65s/it] + Training...: 83% 3651/4393 [4:58:01<37:05, 3.00s/it] + Training...: 83% 3652/4393 [4:58:07<48:59, 3.97s/it] + Training...: 83% 3653/4393 [4:58:13<56:46, 4.60s/it] + Training...: 83% 3654/4393 [4:58:19<1:02:00, 5.03s/it] + Training...: 83% 3655/4393 [4:58:25<1:05:23, 5.32s/it] + Training...: 83% 3656/4393 [4:58:31<1:07:39, 5.51s/it] + Training...: 83% 3657/4393 [4:58:37<1:08:57, 5.62s/it] + Training...: 83% 3658/4393 [4:58:43<1:09:46, 5.70s/it] + Training...: 83% 3659/4393 [4:58:49<1:10:29, 5.76s/it] + Training...: 83% 3660/4393 [4:58:55<1:10:41, 5.79s/it] + Training...: 83% 3661/4393 [4:59:01<1:10:37, 5.79s/it] + Training...: 83% 3662/4393 [4:59:06<1:10:28, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:36:13<17:31:17, 21025.84s/it] + Training...: 83% 3662/4393 [4:59:13<1:10:28, 5.78s/it] + Training...: 83% 3663/4393 [4:59:13<1:13:09, 6.01s/it] + Training...: 83% 3664/4393 [4:59:19<1:12:10, 5.94s/it] + Training...: 83% 3665/4393 [4:59:24<1:11:17, 5.88s/it] + Training...: 83% 3666/4393 [4:59:30<1:10:36, 5.83s/it] + Training...: 83% 3667/4393 [4:59:36<1:10:35, 5.83s/it] + Training...: 83% 3668/4393 [4:59:42<1:09:31, 5.75s/it] + Training...: 84% 3669/4393 [4:59:47<1:08:47, 5.70s/it] + Training...: 84% 3670/4393 [4:59:53<1:07:35, 5.61s/it] + Training...: 84% 3671/4393 [4:59:58<1:06:52, 5.56s/it] + Training...: 84% 3672/4393 [5:00:03<1:06:18, 5.52s/it] + Training...: 84% 3673/4393 [5:00:09<1:05:57, 5.50s/it] + Training...: 84% 3674/4393 [5:00:14<1:05:08, 5.44s/it] + Training...: 84% 3675/4393 [5:00:19<1:04:38, 5.40s/it] + Training...: 84% 3676/4393 [5:00:25<1:03:52, 5.34s/it] + Training...: 84% 3677/4393 [5:00:30<1:03:14, 5.30s/it] + Training...: 84% 3678/4393 [5:00:35<1:02:24, 5.24s/it] + Training...: 84% 3679/4393 [5:00:40<1:01:46, 5.19s/it] + Training...: 84% 3680/4393 [5:00:45<1:00:57, 5.13s/it] + Training...: 84% 3681/4393 [5:00:50<1:00:16, 5.08s/it] + Training...: 84% 3682/4393 [5:00:55<1:00:13, 5.08s/it] + Training...: 84% 3683/4393 [5:01:00<59:38, 5.04s/it]  + Training...: 84% 3684/4393 [5:01:05<58:19, 4.94s/it] + Training...: 84% 3685/4393 [5:01:09<56:59, 4.83s/it] + Training...: 84% 3686/4393 [5:01:14<55:40, 4.73s/it] + Training...: 84% 3687/4393 [5:01:18<54:23, 4.62s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:38:23<17:31:17, 21025.84s/it] + Training...: 84% 3687/4393 [5:01:23<54:23, 4.62s/it] + Training...: 84% 3688/4393 [5:01:23<54:47, 4.66s/it] + Training...: 84% 3689/4393 [5:01:27<52:22, 4.46s/it] + Training...: 84% 3690/4393 [5:01:31<49:55, 4.26s/it] + Training...: 84% 3691/4393 [5:01:34<47:13, 4.04s/it] + Training...: 84% 3692/4393 [5:01:37<44:26, 3.80s/it] + Training...: 84% 3693/4393 [5:01:40<41:33, 3.56s/it] + Training...: 84% 3694/4393 [5:01:43<38:39, 3.32s/it] + Training...: 84% 3695/4393 [5:01:46<35:31, 3.05s/it] + Training...: 84% 3696/4393 [5:01:48<32:27, 2.79s/it] + Training...: 84% 3697/4393 [5:01:50<29:22, 2.53s/it] + Training...: 84% 3698/4393 [5:01:51<26:21, 2.28s/it] + Training...: 84% 3699/4393 [5:01:53<23:10, 2.00s/it] + Training...: 84% 3700/4393 [5:01:54<19:51, 1.72s/it] + Training...: 84% 3701/4393 [5:02:00<35:51, 3.11s/it] + Training...: 84% 3702/4393 [5:02:07<46:46, 4.06s/it] + Training...: 84% 3703/4393 [5:02:13<53:54, 4.69s/it] + Training...: 84% 3704/4393 [5:02:19<58:38, 5.11s/it] + Training...: 84% 3705/4393 [5:02:25<1:01:45, 5.39s/it] + Training...: 84% 3706/4393 [5:02:31<1:03:45, 5.57s/it] + Training...: 84% 3707/4393 [5:02:37<1:05:55, 5.77s/it] + Training...: 84% 3708/4393 [5:02:43<1:06:21, 5.81s/it] + Training...: 84% 3709/4393 [5:02:49<1:06:36, 5.84s/it] + Training...: 84% 3710/4393 [5:02:55<1:06:31, 5.84s/it] + Training...: 84% 3711/4393 [5:03:00<1:06:18, 5.83s/it] + Training...: 84% 3712/4393 [5:03:06<1:05:43, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:40:13<17:31:17, 21025.84s/it] + Training...: 84% 3712/4393 [5:03:13<1:05:43, 5.79s/it] + Training...: 85% 3713/4393 [5:03:13<1:08:07, 6.01s/it] + Training...: 85% 3714/4393 [5:03:18<1:07:10, 5.94s/it] + Training...: 85% 3715/4393 [5:03:24<1:06:12, 5.86s/it] + Training...: 85% 3716/4393 [5:03:30<1:05:29, 5.80s/it] + Training...: 85% 3717/4393 [5:03:35<1:04:55, 5.76s/it] + Training...: 85% 3718/4393 [5:03:41<1:03:59, 5.69s/it] + Training...: 85% 3719/4393 [5:03:47<1:03:17, 5.63s/it] + Training...: 85% 3720/4393 [5:03:52<1:02:34, 5.58s/it] + Training...: 85% 3721/4393 [5:03:57<1:02:07, 5.55s/it] + Training...: 85% 3722/4393 [5:04:03<1:01:43, 5.52s/it] + Training...: 85% 3723/4393 [5:04:08<1:01:28, 5.51s/it] + Training...: 85% 3724/4393 [5:04:14<1:00:49, 5.46s/it] + Training...: 85% 3725/4393 [5:04:19<1:00:11, 5.41s/it] + Training...: 85% 3726/4393 [5:04:24<59:34, 5.36s/it]  + Training...: 85% 3727/4393 [5:04:29<59:06, 5.32s/it] + Training...: 85% 3728/4393 [5:04:35<58:25, 5.27s/it] + Training...: 85% 3729/4393 [5:04:40<57:38, 5.21s/it] + Training...: 85% 3730/4393 [5:04:45<56:56, 5.15s/it] + Training...: 85% 3731/4393 [5:04:50<56:06, 5.08s/it] + Training...: 85% 3732/4393 [5:04:54<55:05, 5.00s/it] + Training...: 85% 3733/4393 [5:04:59<54:05, 4.92s/it] + Training...: 85% 3734/4393 [5:05:04<52:54, 4.82s/it] + Training...: 85% 3735/4393 [5:05:08<51:49, 4.73s/it] + Training...: 85% 3736/4393 [5:05:13<50:35, 4.62s/it] + Training...: 85% 3737/4393 [5:05:17<49:29, 4.53s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:42:22<17:31:17, 21025.84s/it] + Training...: 85% 3737/4393 [5:05:22<49:29, 4.53s/it] + Training...: 85% 3738/4393 [5:05:22<50:09, 4.59s/it] + Training...: 85% 3739/4393 [5:05:26<47:37, 4.37s/it] + Training...: 85% 3740/4393 [5:05:29<45:07, 4.15s/it] + Training...: 85% 3741/4393 [5:05:32<42:23, 3.90s/it] + Training...: 85% 3742/4393 [5:05:36<39:36, 3.65s/it] + Training...: 85% 3743/4393 [5:05:38<36:53, 3.40s/it] + Training...: 85% 3744/4393 [5:05:41<34:09, 3.16s/it] + Training...: 85% 3745/4393 [5:05:43<31:26, 2.91s/it] + Training...: 85% 3746/4393 [5:05:45<28:43, 2.66s/it] + Training...: 85% 3747/4393 [5:05:47<26:06, 2.42s/it] + Training...: 85% 3748/4393 [5:05:49<23:21, 2.17s/it] + Training...: 85% 3749/4393 [5:05:50<20:31, 1.91s/it] + Training...: 85% 3750/4393 [5:05:51<17:37, 1.65s/it] + Training...: 85% 3751/4393 [5:05:57<32:17, 3.02s/it] + Training...: 85% 3752/4393 [5:06:04<42:34, 3.99s/it] + Training...: 85% 3753/4393 [5:06:10<49:16, 4.62s/it] + Training...: 85% 3754/4393 [5:06:16<53:47, 5.05s/it] + Training...: 85% 3755/4393 [5:06:22<56:58, 5.36s/it] + Training...: 85% 3756/4393 [5:06:28<58:44, 5.53s/it] + Training...: 86% 3757/4393 [5:06:34<1:00:00, 5.66s/it] + Training...: 86% 3758/4393 [5:06:40<1:00:31, 5.72s/it] + Training...: 86% 3759/4393 [5:06:46<1:01:03, 5.78s/it] + Training...: 86% 3760/4393 [5:06:51<1:01:07, 5.79s/it] + Training...: 86% 3761/4393 [5:06:57<1:01:07, 5.80s/it] + Training...: 86% 3762/4393 [5:07:03<1:02:00, 5.90s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:44:10<17:31:17, 21025.84s/it] + Training...: 86% 3762/4393 [5:07:10<1:02:00, 5.90s/it] + Training...: 86% 3763/4393 [5:07:10<1:03:54, 6.09s/it] + Training...: 86% 3764/4393 [5:07:16<1:02:41, 5.98s/it] + Training...: 86% 3765/4393 [5:07:21<1:01:44, 5.90s/it] + Training...: 86% 3766/4393 [5:07:27<1:00:31, 5.79s/it] + Training...: 86% 3767/4393 [5:07:32<59:48, 5.73s/it]  + Training...: 86% 3768/4393 [5:07:38<59:04, 5.67s/it] + Training...: 86% 3769/4393 [5:07:43<58:32, 5.63s/it] + Training...: 86% 3770/4393 [5:07:49<58:05, 5.59s/it] + Training...: 86% 3771/4393 [5:07:54<57:27, 5.54s/it] + Training...: 86% 3772/4393 [5:08:00<56:48, 5.49s/it] + Training...: 86% 3773/4393 [5:08:05<56:24, 5.46s/it] + Training...: 86% 3774/4393 [5:08:10<55:52, 5.42s/it] + Training...: 86% 3775/4393 [5:08:16<55:17, 5.37s/it] + Training...: 86% 3776/4393 [5:08:21<55:30, 5.40s/it] + Training...: 86% 3777/4393 [5:08:26<54:51, 5.34s/it] + Training...: 86% 3778/4393 [5:08:32<54:02, 5.27s/it] + Training...: 86% 3779/4393 [5:08:37<53:25, 5.22s/it] + Training...: 86% 3780/4393 [5:08:42<52:32, 5.14s/it] + Training...: 86% 3781/4393 [5:08:46<51:42, 5.07s/it] + Training...: 86% 3782/4393 [5:08:51<50:53, 5.00s/it] + Training...: 86% 3783/4393 [5:08:56<50:18, 4.95s/it] + Training...: 86% 3784/4393 [5:09:01<49:21, 4.86s/it] + Training...: 86% 3785/4393 [5:09:05<48:31, 4.79s/it] + Training...: 86% 3786/4393 [5:09:10<47:28, 4.69s/it] + Training...: 86% 3787/4393 [5:09:14<46:20, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:46:19<17:31:17, 21025.84s/it] + Training...: 86% 3787/4393 [5:09:19<46:20, 4.59s/it] + Training...: 86% 3788/4393 [5:09:19<46:47, 4.64s/it] + Training...: 86% 3789/4393 [5:09:23<44:29, 4.42s/it] + Training...: 86% 3790/4393 [5:09:27<42:03, 4.18s/it] + Training...: 86% 3791/4393 [5:09:30<39:33, 3.94s/it] + Training...: 86% 3792/4393 [5:09:33<37:06, 3.70s/it] + Training...: 86% 3793/4393 [5:09:36<34:44, 3.47s/it] + Training...: 86% 3794/4393 [5:09:39<32:10, 3.22s/it] + Training...: 86% 3795/4393 [5:09:41<29:32, 2.96s/it] + Training...: 86% 3796/4393 [5:09:43<26:55, 2.71s/it] + Training...: 86% 3797/4393 [5:09:45<24:11, 2.44s/it] + Training...: 86% 3798/4393 [5:09:47<21:41, 2.19s/it] + Training...: 86% 3799/4393 [5:09:48<19:00, 1.92s/it] + Training...: 87% 3800/4393 [5:09:49<16:17, 1.65s/it] + Training...: 87% 3801/4393 [5:09:55<30:11, 3.06s/it] + Training...: 87% 3802/4393 [5:10:01<39:41, 4.03s/it] + Training...: 87% 3803/4393 [5:10:08<45:50, 4.66s/it] + Training...: 87% 3804/4393 [5:10:14<49:56, 5.09s/it] + Training...: 87% 3805/4393 [5:10:20<52:37, 5.37s/it] + Training...: 87% 3806/4393 [5:10:26<54:10, 5.54s/it] + Training...: 87% 3807/4393 [5:10:32<55:25, 5.67s/it] + Training...: 87% 3808/4393 [5:10:37<55:49, 5.73s/it] + Training...: 87% 3809/4393 [5:10:44<56:42, 5.83s/it] + Training...: 87% 3810/4393 [5:10:49<56:47, 5.84s/it] + Training...: 87% 3811/4393 [5:10:55<56:47, 5.85s/it] + Training...: 87% 3812/4393 [5:11:01<56:22, 5.82s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:48:08<17:31:17, 21025.84s/it] + Training...: 87% 3812/4393 [5:11:08<56:22, 5.82s/it] + Training...: 87% 3813/4393 [5:11:08<58:31, 6.05s/it] + Training...: 87% 3814/4393 [5:11:13<57:22, 5.95s/it] + Training...: 87% 3815/4393 [5:11:19<56:47, 5.89s/it] + Training...: 87% 3816/4393 [5:11:25<56:03, 5.83s/it] + Training...: 87% 3817/4393 [5:11:30<55:27, 5.78s/it] + Training...: 87% 3818/4393 [5:11:36<55:35, 5.80s/it] + Training...: 87% 3819/4393 [5:11:42<55:08, 5.76s/it] + Training...: 87% 3820/4393 [5:11:47<54:07, 5.67s/it] + Training...: 87% 3821/4393 [5:11:53<53:30, 5.61s/it] + Training...: 87% 3822/4393 [5:11:58<52:42, 5.54s/it] + Training...: 87% 3823/4393 [5:12:04<52:15, 5.50s/it] + Training...: 87% 3824/4393 [5:12:09<51:48, 5.46s/it] + Training...: 87% 3825/4393 [5:12:14<51:14, 5.41s/it] + Training...: 87% 3826/4393 [5:12:20<50:38, 5.36s/it] + Training...: 87% 3827/4393 [5:12:25<50:04, 5.31s/it] + Training...: 87% 3828/4393 [5:12:30<49:18, 5.24s/it] + Training...: 87% 3829/4393 [5:12:35<48:46, 5.19s/it] + Training...: 87% 3830/4393 [5:12:40<48:07, 5.13s/it] + Training...: 87% 3831/4393 [5:12:45<47:26, 5.07s/it] + Training...: 87% 3832/4393 [5:12:50<46:38, 4.99s/it] + Training...: 87% 3833/4393 [5:12:54<45:55, 4.92s/it] + Training...: 87% 3834/4393 [5:12:59<45:09, 4.85s/it] + Training...: 87% 3835/4393 [5:13:04<44:18, 4.76s/it] + Training...: 87% 3836/4393 [5:13:08<43:19, 4.67s/it] + Training...: 87% 3837/4393 [5:13:12<41:58, 4.53s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:50:17<17:31:17, 21025.84s/it] + Training...: 87% 3837/4393 [5:13:17<41:58, 4.53s/it] + Training...: 87% 3838/4393 [5:13:17<42:08, 4.56s/it] + Training...: 87% 3839/4393 [5:13:21<39:58, 4.33s/it] + Training...: 87% 3840/4393 [5:13:24<37:51, 4.11s/it] + Training...: 87% 3841/4393 [5:13:28<35:36, 3.87s/it] + Training...: 87% 3842/4393 [5:13:31<33:32, 3.65s/it] + Training...: 87% 3843/4393 [5:13:34<31:07, 3.40s/it] + Training...: 88% 3844/4393 [5:13:36<28:46, 3.14s/it] + Training...: 88% 3845/4393 [5:13:38<26:21, 2.89s/it] + Training...: 88% 3846/4393 [5:13:40<24:00, 2.63s/it] + Training...: 88% 3847/4393 [5:13:42<21:39, 2.38s/it] + Training...: 88% 3848/4393 [5:13:44<19:21, 2.13s/it] + Training...: 88% 3849/4393 [5:13:45<17:06, 1.89s/it] + Training...: 88% 3850/4393 [5:13:46<14:48, 1.64s/it] + Training...: 88% 3851/4393 [5:13:52<27:14, 3.02s/it] + Training...: 88% 3852/4393 [5:13:59<36:00, 3.99s/it] + Training...: 88% 3853/4393 [5:14:05<41:41, 4.63s/it] + Training...: 88% 3854/4393 [5:14:11<45:25, 5.06s/it] + Training...: 88% 3855/4393 [5:14:17<48:03, 5.36s/it] + Training...: 88% 3856/4393 [5:14:23<49:28, 5.53s/it] + Training...: 88% 3857/4393 [5:14:29<50:28, 5.65s/it] + Training...: 88% 3858/4393 [5:14:35<50:59, 5.72s/it] + Training...: 88% 3859/4393 [5:14:41<51:24, 5.78s/it] + Training...: 88% 3860/4393 [5:14:46<51:23, 5.78s/it] + Training...: 88% 3861/4393 [5:14:52<51:21, 5.79s/it] + Training...: 88% 3862/4393 [5:14:58<51:10, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:52:04<17:31:17, 21025.84s/it] +Step... (41700 | Loss: 0.005744441412389278, Learning Rate: 1.676969441177789e-05, Gradient Norm: 0.1374588906764984) +Step... (41725 | Loss: 0.008426920510828495, Learning Rate: 1.6719191989977844e-05, Gradient Norm: 0.8020673990249634) +Step... (41750 | Loss: 0.0075722914189100266, Learning Rate: 1.6668689568177797e-05, Gradient Norm: 0.32124248147010803) +Step... (41775 | Loss: 0.005043766926974058, Learning Rate: 1.6618179870420136e-05, Gradient Norm: 0.20893269777297974) +Step... (41800 | Loss: 0.010060012340545654, Learning Rate: 1.656767744862009e-05, Gradient Norm: 0.21614883840084076) +Step... (41825 | Loss: 0.014381218701601028, Learning Rate: 1.6517173207830638e-05, Gradient Norm: 0.35426953434944153) +Step... (41850 | Loss: 0.01061459444463253, Learning Rate: 1.646666532906238e-05, Gradient Norm: 0.29452937841415405) +Step... (41875 | Loss: 0.012180848978459835, Learning Rate: 1.641616108827293e-05, Gradient Norm: 0.3592362105846405) +Step... (41900 | Loss: 0.00836265180259943, Learning Rate: 1.6365658666472882e-05, Gradient Norm: 0.22259344160556793) +Step... (41925 | Loss: 0.008231023326516151, Learning Rate: 1.631514896871522e-05, Gradient Norm: 0.20443135499954224) +Step... (41950 | Loss: 0.01289241760969162, Learning Rate: 1.6264646546915174e-05, Gradient Norm: 0.3498441278934479) +Step... (41975 | Loss: 0.0034512323327362537, Learning Rate: 1.6214144125115126e-05, Gradient Norm: 0.13234050571918488) +Step... (42000 | Loss: 0.011287801899015903, Learning Rate: 1.6163634427357465e-05, Gradient Norm: 0.4331746995449066) +Step... (42025 | Loss: 0.003170725191012025, Learning Rate: 1.6113132005557418e-05, Gradient Norm: 0.13511165976524353) +Step... (42050 | Loss: 0.011994440108537674, Learning Rate: 1.606262958375737e-05, Gradient Norm: 0.2843227684497833) +Step... (42075 | Loss: 0.003315770300105214, Learning Rate: 1.601211988599971e-05, Gradient Norm: 0.18491697311401367) +Step... (42100 | Loss: 0.011447119526565075, Learning Rate: 1.5961617464199662e-05, Gradient Norm: 0.23545017838478088) +Step... (42125 | Loss: 0.005759582854807377, Learning Rate: 1.591111322341021e-05, Gradient Norm: 0.2273390293121338) +Step... (42150 | Loss: 0.010285439901053905, Learning Rate: 1.5860605344641954e-05, Gradient Norm: 0.2514696419239044) +Step... (42175 | Loss: 0.004782717674970627, Learning Rate: 1.5810101103852503e-05, Gradient Norm: 0.17328862845897675) +Step... (42200 | Loss: 0.010704944841563702, Learning Rate: 1.5759598682052456e-05, Gradient Norm: 0.26759225130081177) +Step... (42225 | Loss: 0.008894861675798893, Learning Rate: 1.5709088984294794e-05, Gradient Norm: 0.21136222779750824) +Step... (42250 | Loss: 0.014378275722265244, Learning Rate: 1.5658586562494747e-05, Gradient Norm: 0.2818372845649719) +Step... (42275 | Loss: 0.003270916873589158, Learning Rate: 1.56080841406947e-05, Gradient Norm: 0.1778145730495453) +Step... (42300 | Loss: 0.01763617806136608, Learning Rate: 1.555757444293704e-05, Gradient Norm: 0.8628730773925781) +Step... (42325 | Loss: 0.003942512441426516, Learning Rate: 1.550707202113699e-05, Gradient Norm: 0.13781097531318665) +Step... (42350 | Loss: 0.0093576330691576, Learning Rate: 1.545656232337933e-05, Gradient Norm: 0.1558447927236557) +Step... (42375 | Loss: 0.005177142098546028, Learning Rate: 1.5406059901579283e-05, Gradient Norm: 0.19731150567531586) +Step... (42400 | Loss: 0.008837847970426083, Learning Rate: 1.5355557479779236e-05, Gradient Norm: 0.2248029261827469) +Step... (42425 | Loss: 0.004063450265675783, Learning Rate: 1.5305047782021575e-05, Gradient Norm: 0.15096257627010345) +Step... (42450 | Loss: 0.0156438909471035, Learning Rate: 1.5254544450726826e-05, Gradient Norm: 0.5371332168579102) +Step... (42475 | Loss: 0.004050035495311022, Learning Rate: 1.5204042028926779e-05, Gradient Norm: 0.2783503830432892) +Step... (42500 | Loss: 0.008893079124391079, Learning Rate: 1.5153533240663819e-05, Gradient Norm: 0.16905491054058075) +Step... (42525 | Loss: 0.0038313837721943855, Learning Rate: 1.510302990936907e-05, Gradient Norm: 0.13977232575416565) +Step... (42550 Training...: 88% 3862/4393 [5:15:05<51:10, 5.78s/it] + Training...: 88% 3863/4393 [5:15:05<53:14, 6.03s/it]| Loss: 0.008676256984472275, Learning Rate: 1.5052526578074321e-05, Gradient Norm: 0.1817447394132614) +Step... (42575 | Loss: 0.005087659694254398, Learning Rate: 1.5002017789811362e-05, Gradient Norm: 0.44567009806632996) +Step... (42600 | Loss: 0.013834933750331402, Learning Rate: 1.4951514458516613e-05, Gradient Norm: 0.46296262741088867) +Step... (42625 | Loss: 0.01469030138105154, Learning Rate: 1.4901012036716565e-05, Gradient Norm: 0.6844345331192017) +Step... (42650 | Loss: 0.005226873327046633, Learning Rate: 1.4850503248453606e-05, Gradient Norm: 0.18194365501403809) +Step... (42675 | Loss: 0.010348882526159286, Learning Rate: 1.4799999917158857e-05, Gradient Norm: 4.603023529052734) +Step... (42700 | Loss: 0.00900035910308361, Learning Rate: 1.4749496585864108e-05, Gradient Norm: 0.22959791123867035) +Step... (42725 | Loss: 0.004127497784793377, Learning Rate: 1.4698987797601148e-05, Gradient Norm: 0.16160161793231964) +Step... (42750 | Loss: 0.009940936230123043, Learning Rate: 1.46484844663064e-05, Gradient Norm: 0.2361784130334854) +Step... (42775 | Loss: 0.009671159088611603, Learning Rate: 1.4597982044506352e-05, Gradient Norm: 0.24877315759658813) +Step... (42800 | Loss: 0.010518192313611507, Learning Rate: 1.4547473256243393e-05, Gradient Norm: 0.2851322591304779) +Step... (42825 | Loss: 0.010680184699594975, Learning Rate: 1.4496969924948644e-05, Gradient Norm: 0.5760316848754883) +Step... (42850 | Loss: 0.008422491140663624, Learning Rate: 1.4446466593653895e-05, Gradient Norm: 0.2787226438522339) +Step... (42875 | Loss: 0.013169726356863976, Learning Rate: 1.4395957805390935e-05, Gradient Norm: 0.2936961054801941) +Step... (42900 | Loss: 0.007032770197838545, Learning Rate: 1.4345454474096186e-05, Gradient Norm: 0.20587222278118134) +Step... (42925 | Loss: 0.004653589334338903, Learning Rate: 1.429495205229614e-05, Gradient Norm: 0.18729034066200256) +Step... (42950 | Loss: 0.005974963773041964, Learning Rate: 1.424444326403318e-05, Gradient Norm: 0.1742350161075592) +Step... (42975 | Loss: 0.0030157752335071564, Learning Rate: 1.419393993273843e-05, Gradient Norm: 0.09114522486925125) +Step... (43000 | Loss: 0.017072968184947968, Learning Rate: 1.4143436601443682e-05, Gradient Norm: 0.344232439994812) +Step... (43025 | Loss: 0.007778532337397337, Learning Rate: 1.4092927813180722e-05, Gradient Norm: 0.24894601106643677) +Step... (43050 | Loss: 0.005137579515576363, Learning Rate: 1.4042424481885973e-05, Gradient Norm: 0.15516820549964905) +Step... (43075 | Loss: 0.008852087892591953, Learning Rate: 1.3991922060085926e-05, Gradient Norm: 0.2543483078479767) +Step... (43100 | Loss: 0.005538458935916424, Learning Rate: 1.3941412362328265e-05, Gradient Norm: 0.20741888880729675) +Step... (43125 | Loss: 0.009585311636328697, Learning Rate: 1.3890909940528218e-05, Gradient Norm: 0.32343360781669617) +Step... (43150 | Loss: 0.009101358242332935, Learning Rate: 1.3840406609233469e-05, Gradient Norm: 0.1852715015411377) +Step... (43175 | Loss: 0.0038933581672608852, Learning Rate: 1.378989782097051e-05, Gradient Norm: 0.200323686003685) +Step... (43200 | Loss: 0.006398520432412624, Learning Rate: 1.373939448967576e-05, Gradient Norm: 0.3812198340892792) +Step... (43225 | Loss: 0.00649980129674077, Learning Rate: 1.3688892067875713e-05, Gradient Norm: 0.2510472238063812) +Step... (43250 | Loss: 0.007147450000047684, Learning Rate: 1.3638382370118052e-05, Gradient Norm: 0.2402760088443756) +Step... (43275 | Loss: 0.005947265308350325, Learning Rate: 1.3587879948318005e-05, Gradient Norm: 0.21199539303779602) +Step... (43300 | Loss: 0.01116273459047079, Learning Rate: 1.3537376617023256e-05, Gradient Norm: 0.1889847368001938) +Step... (43325 | Loss: 0.009462771005928516, Learning Rate: 1.3486867828760296e-05, Gradient Norm: 0.4412449598312378) +Step... (43350 | Loss: 0.01658274605870247, Learning Rate: 1.3436364497465547e-05, Gradient Norm: 0.24183489382266998) +Step... (43375 | Loss: 0.005811340641230345, Learning Rate: 1.3385855709202588e-05, Gradient Norm: 0.13360491394996643) + + Training...: 88% 3864/4393 [5:15:11<53:02, 6.02s/it] + Training...: 88% 3865/4393 [5:15:16<52:14, 5.94s/it] + Training...: 88% 3866/4393 [5:15:22<51:26, 5.86s/it] + Training...: 88% 3867/4393 [5:15:28<51:14, 5.85s/it] + Training...: 88% 3868/4393 [5:15:33<50:24, 5.76s/it] + Training...: 88% 3869/4393 [5:15:39<49:53, 5.71s/it] + Training...: 88% 3870/4393 [5:15:44<49:17, 5.65s/it] + Training...: 88% 3871/4393 [5:15:50<48:45, 5.61s/it] + Training...: 88% 3872/4393 [5:15:55<48:08, 5.54s/it] + Training...: 88% 3873/4393 [5:16:01<47:30, 5.48s/it] + Training...: 88% 3874/4393 [5:16:06<46:53, 5.42s/it] + Training...: 88% 3875/4393 [5:16:11<46:29, 5.39s/it] + Training...: 88% 3876/4393 [5:16:16<45:52, 5.32s/it] + Training...: 88% 3877/4393 [5:16:22<45:22, 5.28s/it] + Training...: 88% 3878/4393 [5:16:27<44:43, 5.21s/it] + Training...: 88% 3879/4393 [5:16:32<44:13, 5.16s/it] + Training...: 88% 3880/4393 [5:16:37<43:41, 5.11s/it] + Training...: 88% 3881/4393 [5:16:42<43:10, 5.06s/it] + Training...: 88% 3882/4393 [5:16:46<42:28, 4.99s/it] + Training...: 88% 3883/4393 [5:16:51<41:48, 4.92s/it] + Training...: 88% 3884/4393 [5:16:56<41:03, 4.84s/it] + Training...: 88% 3885/4393 [5:17:00<40:07, 4.74s/it] + Training...: 88% 3886/4393 [5:17:05<39:16, 4.65s/it] + Training...: 88% 3887/4393 [5:17:09<38:10, 4.53s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:54:14<17:31:17, 21025.84s/it] + Training...: 88% 3887/4393 [5:17:14<38:10, 4.53s/it] + Training...: 89% 3888/4393 [5:17:14<38:32, 4.58s/it] + Training...: 89% 3889/4393 [5:17:18<36:33, 4.35s/it] + Training...: 89% 3890/4393 [5:17:21<34:30, 4.12s/it] + Training...: 89% 3891/4393 [5:17:24<32:18, 3.86s/it] + Training...: 89% 3892/4393 [5:17:27<30:06, 3.61s/it] + Training...: 89% 3893/4393 [5:17:30<28:07, 3.37s/it] + Training...: 89% 3894/4393 [5:17:33<26:13, 3.15s/it] + Training...: 89% 3895/4393 [5:17:35<24:14, 2.92s/it] + Training...: 89% 3896/4393 [5:17:37<22:09, 2.67s/it] + Training...: 89% 3897/4393 [5:17:39<20:05, 2.43s/it] + Training...: 89% 3898/4393 [5:17:41<17:59, 2.18s/it] + Training...: 89% 3899/4393 [5:17:42<15:53, 1.93s/it] + Training...: 89% 3900/4393 [5:17:43<13:36, 1.66s/it] + Training...: 89% 3901/4393 [5:17:49<24:54, 3.04s/it] + Training...: 89% 3902/4393 [5:17:56<32:57, 4.03s/it] + Training...: 89% 3903/4393 [5:18:02<38:02, 4.66s/it] + Training...: 89% 3904/4393 [5:18:08<41:10, 5.05s/it] + Training...: 89% 3905/4393 [5:18:14<43:24, 5.34s/it] + Training...: 89% 3906/4393 [5:18:20<45:05, 5.56s/it] + Training...: 89% 3907/4393 [5:18:26<46:07, 5.70s/it] + Training...: 89% 3908/4393 [5:18:32<46:36, 5.77s/it] + Training...: 89% 3909/4393 [5:18:38<46:51, 5.81s/it] + Training...: 89% 3910/4393 [5:18:44<46:51, 5.82s/it] + Training...: 89% 3911/4393 [5:18:50<46:48, 5.83s/it] + Training...: 89% 3912/4393 [5:18:55<46:30, 5.80s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:56:02<17:31:17, 21025.84s/it] + Training...: 89% 3912/4393 [5:19:02<46:30, 5.80s/it] + Training...: 89% 3913/4393 [5:19:02<48:36, 6.08s/it] + Training...: 89% 3914/4393 [5:19:08<47:52, 6.00s/it] + Training...: 89% 3915/4393 [5:19:14<47:08, 5.92s/it] + Training...: 89% 3916/4393 [5:19:19<46:24, 5.84s/it] + Training...: 89% 3917/4393 [5:19:25<45:48, 5.77s/it] + Training...: 89% 3918/4393 [5:19:30<45:07, 5.70s/it] + Training...: 89% 3919/4393 [5:19:36<44:30, 5.63s/it] + Training...: 89% 3920/4393 [5:19:41<44:12, 5.61s/it] + Training...: 89% 3921/4393 [5:19:47<43:47, 5.57s/it] + Training...: 89% 3922/4393 [5:19:52<43:11, 5.50s/it] + Training...: 89% 3923/4393 [5:19:58<42:44, 5.46s/it] + Training...: 89% 3924/4393 [5:20:03<42:18, 5.41s/it] + Training...: 89% 3925/4393 [5:20:08<41:48, 5.36s/it] + Training...: 89% 3926/4393 [5:20:13<41:23, 5.32s/it] + Training...: 89% 3927/4393 [5:20:19<41:07, 5.29s/it] + Training...: 89% 3928/4393 [5:20:24<40:34, 5.23s/it] + Training...: 89% 3929/4393 [5:20:29<40:04, 5.18s/it] + Training...: 89% 3930/4393 [5:20:34<39:32, 5.12s/it] + Training...: 89% 3931/4393 [5:20:39<39:13, 5.09s/it] + Training...: 90% 3932/4393 [5:20:44<38:57, 5.07s/it] + Training...: 90% 3933/4393 [5:20:49<38:18, 5.00s/it] + Training...: 90% 3934/4393 [5:20:53<37:30, 4.90s/it] + Training...: 90% 3935/4393 [5:20:58<36:40, 4.80s/it] + Training...: 90% 3936/4393 [5:21:02<35:44, 4.69s/it] + Training...: 90% 3937/4393 [5:21:07<34:45, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [57:58:11<17:31:17, 21025.84s/it] + Training...: 90% 3937/4393 [5:21:11<34:45, 4.57s/it] + Training...: 90% 3938/4393 [5:21:11<34:57, 4.61s/it] + Training...: 90% 3939/4393 [5:21:15<33:14, 4.39s/it] + Training...: 90% 3940/4393 [5:21:19<31:31, 4.18s/it] + Training...: 90% 3941/4393 [5:21:22<29:40, 3.94s/it] + Training...: 90% 3942/4393 [5:21:25<27:42, 3.69s/it] + Training...: 90% 3943/4393 [5:21:28<25:42, 3.43s/it] + Training...: 90% 3944/4393 [5:21:31<23:50, 3.19s/it] + Training...: 90% 3945/4393 [5:21:33<21:47, 2.92s/it] + Training...: 90% 3946/4393 [5:21:35<19:43, 2.65s/it] + Training...: 90% 3947/4393 [5:21:37<17:40, 2.38s/it] + Training...: 90% 3948/4393 [5:21:38<15:42, 2.12s/it] + Training...: 90% 3949/4393 [5:21:40<13:51, 1.87s/it] + Training...: 90% 3950/4393 [5:21:41<11:58, 1.62s/it] + Training...: 90% 3951/4393 [5:21:47<22:14, 3.02s/it] + Training...: 90% 3952/4393 [5:21:53<29:25, 4.00s/it] + Training...: 90% 3953/4393 [5:21:59<34:03, 4.64s/it] + Training...: 90% 3954/4393 [5:22:05<37:10, 5.08s/it] + Training...: 90% 3955/4393 [5:22:11<39:14, 5.38s/it] + Training...: 90% 3956/4393 [5:22:18<41:13, 5.66s/it] + Training...: 90% 3957/4393 [5:22:24<41:58, 5.78s/it] + Training...: 90% 3958/4393 [5:22:30<42:07, 5.81s/it] + Training...: 90% 3959/4393 [5:22:36<42:21, 5.85s/it] + Training...: 90% 3960/4393 [5:22:42<42:07, 5.84s/it] + Training...: 90% 3961/4393 [5:22:47<42:03, 5.84s/it] + Training...: 90% 3962/4393 [5:22:53<41:48, 5.82s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:00:00<17:31:17, 21025.84s/it] + Training...: 90% 3962/4393 [5:23:00<41:48, 5.82s/it] + Training...: 90% 3963/4393 [5:23:00<43:25, 6.06s/it] + Training...: 90% 3964/4393 [5:23:05<42:36, 5.96s/it] + Training...: 90% 3965/4393 [5:23:11<41:57, 5.88s/it] + Training...: 90% 3966/4393 [5:23:17<41:27, 5.82s/it] + Training...: 90% 3967/4393 [5:23:22<40:57, 5.77s/it] + Training...: 90% 3968/4393 [5:23:28<40:21, 5.70s/it] + Training...: 90% 3969/4393 [5:23:34<39:51, 5.64s/it] + Training...: 90% 3970/4393 [5:23:39<39:30, 5.60s/it] + Training...: 90% 3971/4393 [5:23:45<39:07, 5.56s/it] + Training...: 90% 3972/4393 [5:23:50<38:39, 5.51s/it] + Training...: 90% 3973/4393 [5:23:55<38:17, 5.47s/it] + Training...: 90% 3974/4393 [5:24:01<37:55, 5.43s/it] + Training...: 90% 3975/4393 [5:24:06<37:30, 5.38s/it] + Training...: 91% 3976/4393 [5:24:11<37:04, 5.33s/it] + Training...: 91% 3977/4393 [5:24:16<36:38, 5.28s/it] + Training...: 91% 3978/4393 [5:24:21<36:18, 5.25s/it] + Training...: 91% 3979/4393 [5:24:26<35:46, 5.18s/it] + Training...: 91% 3980/4393 [5:24:31<35:17, 5.13s/it] + Training...: 91% 3981/4393 [5:24:36<34:46, 5.06s/it] + Training...: 91% 3982/4393 [5:24:41<34:10, 4.99s/it] + Training...: 91% 3983/4393 [5:24:46<33:47, 4.94s/it] + Training...: 91% 3984/4393 [5:24:51<33:24, 4.90s/it] + Training...: 91% 3985/4393 [5:24:55<32:46, 4.82s/it] + Training...: 91% 3986/4393 [5:25:00<31:58, 4.71s/it] + Training...: 91% 3987/4393 [5:25:04<31:01, 4.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:02:09<17:31:17, 21025.84s/it] + Training...: 91% 3987/4393 [5:25:09<31:01, 4.59s/it] + Training...: 91% 3988/4393 [5:25:09<31:07, 4.61s/it] + Training...: 91% 3989/4393 [5:25:13<29:30, 4.38s/it] + Training...: 91% 3990/4393 [5:25:16<27:59, 4.17s/it] + Training...: 91% 3991/4393 [5:25:20<26:23, 3.94s/it] + Training...: 91% 3992/4393 [5:25:23<24:45, 3.70s/it] + Training...: 91% 3993/4393 [5:25:26<23:03, 3.46s/it] + Training...: 91% 3994/4393 [5:25:29<21:25, 3.22s/it] + Training...: 91% 3995/4393 [5:25:31<19:42, 2.97s/it] + Training...: 91% 3996/4393 [5:25:33<17:58, 2.72s/it] + Training...: 91% 3997/4393 [5:25:35<16:09, 2.45s/it] + Training...: 91% 3998/4393 [5:25:36<14:21, 2.18s/it] + Training...: 91% 3999/4393 [5:25:38<12:34, 1.91s/it] + Training...: 91% 4000/4393 [5:25:39<10:46, 1.64s/it] + Training...: 91% 4001/4393 [5:25:45<19:48, 3.03s/it] + Training...: 91% 4002/4393 [5:25:51<26:01, 3.99s/it] + Training...: 91% 4003/4393 [5:25:57<30:13, 4.65s/it] + Training...: 91% 4004/4393 [5:26:03<32:51, 5.07s/it] + Training...: 91% 4005/4393 [5:26:10<34:39, 5.36s/it] + Training...: 91% 4006/4393 [5:26:15<35:44, 5.54s/it] + Training...: 91% 4007/4393 [5:26:21<36:24, 5.66s/it] + Training...: 91% 4008/4393 [5:26:27<37:09, 5.79s/it] + Training...: 91% 4009/4393 [5:26:33<37:19, 5.83s/it] + Training...: 91% 4010/4393 [5:26:39<37:13, 5.83s/it] + Training...: 91% 4011/4393 [5:26:45<37:05, 5.83s/it] + Training...: 91% 4012/4393 [5:26:51<36:54, 5.81s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:03:57<17:31:17, 21025.84s/it] + Training...: 91% 4012/4393 [5:26:57<36:54, 5.81s/it] + Training...: 91% 4013/4393 [5:26:57<38:10, 6.03s/it] + Training...: 91% 4014/4393 [5:27:03<37:29, 5.94s/it] + Training...: 91% 4015/4393 [5:27:09<36:50, 5.85s/it] + Training...: 91% 4016/4393 [5:27:14<36:24, 5.79s/it] + Training...: 91% 4017/4393 [5:27:20<36:10, 5.77s/it] + Training...: 91% 4018/4393 [5:27:26<35:41, 5.71s/it] + Training...: 91% 4019/4393 [5:27:31<35:19, 5.67s/it] + Training...: 92% 4020/4393 [5:27:37<34:55, 5.62s/it] + Training...: 92% 4021/4393 [5:27:42<34:34, 5.58s/it] + Training...: 92% 4022/4393 [5:27:48<34:19, 5.55s/it] + Training...: 92% 4023/4393 [5:27:53<34:23, 5.58s/it] + Training...: 92% 4024/4393 [5:27:59<33:51, 5.51s/it] + Training...: 92% 4025/4393 [5:28:04<33:20, 5.44s/it] + Training...: 92% 4026/4393 [5:28:09<32:50, 5.37s/it] + Training...: 92% 4027/4393 [5:28:14<32:24, 5.31s/it] + Training...: 92% 4028/4393 [5:28:20<31:59, 5.26s/it] + Training...: 92% 4029/4393 [5:28:25<31:37, 5.21s/it] + Training...: 92% 4030/4393 [5:28:30<31:11, 5.16s/it] + Training...: 92% 4031/4393 [5:28:35<30:45, 5.10s/it] + Training...: 92% 4032/4393 [5:28:39<30:10, 5.02s/it] + Training...: 92% 4033/4393 [5:28:44<29:41, 4.95s/it] + Training...: 92% 4034/4393 [5:28:49<29:07, 4.87s/it] + Training...: 92% 4035/4393 [5:28:54<28:35, 4.79s/it] + Training...: 92% 4036/4393 [5:28:58<27:50, 4.68s/it] + Training...: 92% 4037/4393 [5:29:02<27:10, 4.58s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:06:07<17:31:17, 21025.84s/it] + Training...: 92% 4037/4393 [5:29:07<27:10, 4.58s/it] + Training...: 92% 4038/4393 [5:29:07<27:27, 4.64s/it] + Training...: 92% 4039/4393 [5:29:11<26:12, 4.44s/it] + Training...: 92% 4040/4393 [5:29:15<24:58, 4.24s/it] + Training...: 92% 4041/4393 [5:29:18<23:24, 3.99s/it] + Training...: 92% 4042/4393 [5:29:21<21:52, 3.74s/it] + Training...: 92% 4043/4393 [5:29:24<20:18, 3.48s/it] + Training...: 92% 4044/4393 [5:29:27<18:51, 3.24s/it] + Training...: 92% 4045/4393 [5:29:29<17:20, 2.99s/it] + Training...: 92% 4046/4393 [5:29:32<16:29, 2.85s/it] + Training...: 92% 4047/4393 [5:29:34<14:42, 2.55s/it] + Training...: 92% 4048/4393 [5:29:35<13:09, 2.29s/it] + Training...: 92% 4049/4393 [5:29:37<11:29, 2.01s/it] + Training...: 92% 4050/4393 [5:29:38<09:47, 1.71s/it] + Training...: 92% 4051/4393 [5:29:44<17:35, 3.09s/it] + Training...: 92% 4052/4393 [5:29:51<23:16, 4.10s/it] + Training...: 92% 4053/4393 [5:29:57<26:37, 4.70s/it] + Training...: 92% 4054/4393 [5:30:03<28:40, 5.07s/it] + Training...: 92% 4055/4393 [5:30:09<30:07, 5.35s/it] + Training...: 92% 4056/4393 [5:30:15<31:08, 5.54s/it] + Training...: 92% 4057/4393 [5:30:21<31:49, 5.68s/it] + Training...: 92% 4058/4393 [5:30:27<32:10, 5.76s/it] + Training...: 92% 4059/4393 [5:30:32<32:18, 5.80s/it] + Training...: 92% 4060/4393 [5:30:38<32:17, 5.82s/it] + Training...: 92% 4061/4393 [5:30:44<32:18, 5.84s/it] + Training...: 92% 4062/4393 [5:30:50<32:09, 5.83s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:07:57<17:31:17, 21025.84s/it] + Training...: 92% 4062/4393 [5:30:57<32:09, 5.83s/it] + Training...: 92% 4063/4393 [5:30:57<33:43, 6.13s/it] + Training...: 93% 4064/4393 [5:31:02<32:48, 5.98s/it] + Training...: 93% 4065/4393 [5:31:08<32:15, 5.90s/it] + Training...: 93% 4066/4393 [5:31:14<31:38, 5.81s/it] + Training...: 93% 4067/4393 [5:31:19<31:08, 5.73s/it] + Training...: 93% 4068/4393 [5:31:25<30:46, 5.68s/it] + Training...: 93% 4069/4393 [5:31:30<30:21, 5.62s/it] + Training...: 93% 4070/4393 [5:31:36<30:03, 5.58s/it] + Training...: 93% 4071/4393 [5:31:41<29:45, 5.54s/it] + Training...: 93% 4072/4393 [5:31:47<29:27, 5.51s/it] + Training...: 93% 4073/4393 [5:31:52<29:09, 5.47s/it] + Training...: 93% 4074/4393 [5:31:57<28:47, 5.42s/it] + Training...: 93% 4075/4393 [5:32:03<28:25, 5.36s/it] + Training...: 93% 4076/4393 [5:32:08<27:58, 5.29s/it] + Training...: 93% 4077/4393 [5:32:13<27:44, 5.27s/it] + Training...: 93% 4078/4393 [5:32:18<27:25, 5.22s/it] + Training...: 93% 4079/4393 [5:32:23<27:06, 5.18s/it] + Training...: 93% 4080/4393 [5:32:28<26:42, 5.12s/it] + Training...: 93% 4081/4393 [5:32:33<26:16, 5.05s/it] + Training...: 93% 4082/4393 [5:32:38<26:08, 5.04s/it] + Training...: 93% 4083/4393 [5:32:43<25:47, 4.99s/it] + Training...: 93% 4084/4393 [5:32:48<25:11, 4.89s/it] + Training...: 93% 4085/4393 [5:32:52<24:37, 4.80s/it] + Training...: 93% 4086/4393 [5:32:57<24:05, 4.71s/it] + Training...: 93% 4087/4393 [5:33:01<23:17, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:10:06<17:31:17, 21025.84s/it] + Training...: 93% 4087/4393 [5:33:06<23:17, 4.57s/it] + Training...: 93% 4088/4393 [5:33:06<23:27, 4.62s/it] + Training...: 93% 4089/4393 [5:33:10<22:18, 4.40s/it] + Training...: 93% 4090/4393 [5:33:13<21:11, 4.20s/it] + Training...: 93% 4091/4393 [5:33:17<19:56, 3.96s/it] + Training...: 93% 4092/4393 [5:33:20<18:38, 3.72s/it] + Training...: 93% 4093/4393 [5:33:23<17:16, 3.46s/it] + Training...: 93% 4094/4393 [5:33:25<15:57, 3.20s/it] + Training...: 93% 4095/4393 [5:33:28<14:42, 2.96s/it] + Training...: 93% 4096/4393 [5:33:30<13:24, 2.71s/it] + Training...: 93% 4097/4393 [5:33:32<12:06, 2.45s/it] + Training...: 93% 4098/4393 [5:33:33<10:51, 2.21s/it] + Training...: 93% 4099/4393 [5:33:35<09:32, 1.95s/it] + Training...: 93% 4100/4393 [5:33:36<08:14, 1.69s/it] + Training...: 93% 4101/4393 [5:33:42<14:49, 3.05s/it] + Training...: 93% 4102/4393 [5:33:48<19:32, 4.03s/it] + Training...: 93% 4103/4393 [5:33:54<22:26, 4.64s/it] + Training...: 93% 4104/4393 [5:34:00<24:27, 5.08s/it] + Training...: 93% 4105/4393 [5:34:06<25:44, 5.36s/it] + Training...: 93% 4106/4393 [5:34:12<26:30, 5.54s/it] + Training...: 93% 4107/4393 [5:34:18<27:09, 5.70s/it] + Training...: 94% 4108/4393 [5:34:25<27:42, 5.83s/it] + Training...: 94% 4109/4393 [5:34:30<27:41, 5.85s/it] + Training...: 94% 4110/4393 [5:34:36<27:32, 5.84s/it] + Training...: 94% 4111/4393 [5:34:42<27:26, 5.84s/it] + Training...: 94% 4112/4393 [5:34:48<27:11, 5.80s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:11:54<17:31:17, 21025.84s/it] + Training...: 94% 4112/4393 [5:34:54<27:11, 5.80s/it] + Training...: 94% 4113/4393 [5:34:54<28:06, 6.02s/it] + Training...: 94% 4114/4393 [5:35:00<27:33, 5.93s/it] + Training...: 94% 4115/4393 [5:35:06<27:09, 5.86s/it] + Training...: 94% 4116/4393 [5:35:11<26:41, 5.78s/it] + Training...: 94% 4117/4393 [5:35:17<26:18, 5.72s/it] + Training...: 94% 4118/4393 [5:35:22<25:51, 5.64s/it] + Training...: 94% 4119/4393 [5:35:28<25:36, 5.61s/it] + Training...: 94% 4120/4393 [5:35:33<25:16, 5.56s/it] + Training...: 94% 4121/4393 [5:35:39<25:10, 5.55s/it] + Training...: 94% 4122/4393 [5:35:44<25:03, 5.55s/it] + Training...: 94% 4123/4393 [5:35:50<24:42, 5.49s/it] + Training...: 94% 4124/4393 [5:35:55<24:21, 5.43s/it] + Training...: 94% 4125/4393 [5:36:01<24:13, 5.42s/it] + Training...: 94% 4126/4393 [5:36:06<23:55, 5.38s/it] + Training...: 94% 4127/4393 [5:36:11<23:31, 5.31s/it] + Training...: 94% 4128/4393 [5:36:16<23:06, 5.23s/it] + Training...: 94% 4129/4393 [5:36:21<22:52, 5.20s/it] + Training...: 94% 4130/4393 [5:36:26<22:26, 5.12s/it] + Training...: 94% 4131/4393 [5:36:31<22:09, 5.08s/it] + Training...: 94% 4132/4393 [5:36:36<21:47, 5.01s/it] + Training...: 94% 4133/4393 [5:36:41<21:22, 4.93s/it] + Training...: 94% 4134/4393 [5:36:45<20:48, 4.82s/it] + Training...: 94% 4135/4393 [5:36:50<20:18, 4.72s/it] + Training...: 94% 4136/4393 [5:36:54<19:47, 4.62s/it] + Training...: 94% 4137/4393 [5:36:58<19:15, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:14:03<17:31:17, 21025.84s/it] + Training...: 94% 4137/4393 [5:37:03<19:15, 4.51s/it] + Training...: 94% 4138/4393 [5:37:03<19:21, 4.56s/it] + Training...: 94% 4139/4393 [5:37:07<18:27, 4.36s/it] + Training...: 94% 4140/4393 [5:37:11<17:25, 4.13s/it] + Training...: 94% 4141/4393 [5:37:14<16:19, 3.89s/it] + Training...: 94% 4142/4393 [5:37:17<15:07, 3.62s/it] + Training...: 94% 4143/4393 [5:37:20<14:00, 3.36s/it] + Training...: 94% 4144/4393 [5:37:22<12:53, 3.11s/it] + Training...: 94% 4145/4393 [5:37:24<11:47, 2.85s/it] + Training...: 94% 4146/4393 [5:37:26<10:43, 2.61s/it] + Training...: 94% 4147/4393 [5:37:28<09:40, 2.36s/it] + Training...: 94% 4148/4393 [5:37:30<08:36, 2.11s/it] + Training...: 94% 4149/4393 [5:37:31<07:31, 1.85s/it] + Training...: 94% 4150/4393 [5:37:32<06:28, 1.60s/it] + Training...: 94% 4151/4393 [5:37:38<12:21, 3.06s/it] + Training...: 95% 4152/4393 [5:37:45<16:18, 4.06s/it] + Training...: 95% 4153/4393 [5:37:51<18:43, 4.68s/it] + Training...: 95% 4154/4393 [5:37:57<20:14, 5.08s/it] + Training...: 95% 4155/4393 [5:38:03<21:14, 5.35s/it] + Training...: 95% 4156/4393 [5:38:09<21:53, 5.54s/it] + Training...: 95% 4157/4393 [5:38:15<22:17, 5.67s/it] + Training...: 95% 4158/4393 [5:38:21<22:23, 5.72s/it] + Training...: 95% 4159/4393 [5:38:27<22:29, 5.77s/it] + Training...: 95% 4160/4393 [5:38:32<22:25, 5.78s/it] + Training...: 95% 4161/4393 [5:38:38<22:21, 5.78s/it] + Training...: 95% 4162/4393 [5:38:44<22:15, 5.78s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:15:50<17:31:17, 21025.84s/it] + Training...: 95% 4162/4393 [5:38:51<22:15, 5.78s/it] + Training...: 95% 4163/4393 [5:38:51<23:03, 6.01s/it] + Training...: 95% 4164/4393 [5:38:56<22:31, 5.90s/it] + Training...: 95% 4165/4393 [5:39:02<22:13, 5.85s/it] + Training...: 95% 4166/4393 [5:39:08<21:59, 5.81s/it] + Training...: 95% 4167/4393 [5:39:13<21:47, 5.79s/it] + Training...: 95% 4168/4393 [5:39:19<21:25, 5.71s/it] + Training...: 95% 4169/4393 [5:39:24<21:02, 5.64s/it] + Training...: 95% 4170/4393 [5:39:30<20:46, 5.59s/it] + Training...: 95% 4171/4393 [5:39:35<20:35, 5.57s/it] + Training...: 95% 4172/4393 [5:39:41<20:17, 5.51s/it] + Training...: 95% 4173/4393 [5:39:46<20:05, 5.48s/it] + Training...: 95% 4174/4393 [5:39:52<19:55, 5.46s/it] + Training...: 95% 4175/4393 [5:39:57<19:39, 5.41s/it] + Training...: 95% 4176/4393 [5:40:02<19:28, 5.39s/it] + Training...: 95% 4177/4393 [5:40:08<19:23, 5.39s/it] + Training...: 95% 4178/4393 [5:40:13<19:03, 5.32s/it] + Training...: 95% 4179/4393 [5:40:18<18:44, 5.26s/it] + Training...: 95% 4180/4393 [5:40:23<18:23, 5.18s/it] + Training...: 95% 4181/4393 [5:40:28<18:05, 5.12s/it] + Training...: 95% 4182/4393 [5:40:33<17:45, 5.05s/it] + Training...: 95% 4183/4393 [5:40:38<17:28, 4.99s/it] + Training...: 95% 4184/4393 [5:40:42<17:03, 4.90s/it] + Training...: 95% 4185/4393 [5:40:47<16:42, 4.82s/it] + Training...: 95% 4186/4393 [5:40:51<16:18, 4.73s/it] + Training...: 95% 4187/4393 [5:40:56<15:53, 4.63s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:18:01<17:31:17, 21025.84s/it] + Training...: 95% 4187/4393 [5:41:01<15:53, 4.63s/it] + Training...: 95% 4188/4393 [5:41:01<16:01, 4.69s/it] + Training...: 95% 4189/4393 [5:41:05<15:16, 4.49s/it] + Training...: 95% 4190/4393 [5:41:08<14:31, 4.29s/it] + Training...: 95% 4191/4393 [5:41:12<13:45, 4.08s/it] + Training...: 95% 4192/4393 [5:41:15<12:52, 3.84s/it] + Training...: 95% 4193/4393 [5:41:18<11:55, 3.58s/it] + Training...: 95% 4194/4393 [5:41:21<10:59, 3.31s/it] + Training...: 95% 4195/4393 [5:41:23<10:01, 3.04s/it] + Training...: 96% 4196/4393 [5:41:26<09:08, 2.79s/it] + Training...: 96% 4197/4393 [5:41:28<08:15, 2.53s/it] + Training...: 96% 4198/4393 [5:41:29<07:21, 2.26s/it] + Training...: 96% 4199/4393 [5:41:31<06:27, 2.00s/it] + Training...: 96% 4200/4393 [5:41:32<05:33, 1.73s/it] + Training...: 96% 4201/4393 [5:41:38<09:55, 3.10s/it] + Training...: 96% 4202/4393 [5:41:44<12:53, 4.05s/it] + Training...: 96% 4203/4393 [5:41:50<14:48, 4.68s/it] + Training...: 96% 4204/4393 [5:41:56<16:04, 5.11s/it] + Training...: 96% 4205/4393 [5:42:03<17:08, 5.47s/it] + Training...: 96% 4206/4393 [5:42:09<17:31, 5.62s/it] + Training...: 96% 4207/4393 [5:42:15<17:41, 5.71s/it] + Training...: 96% 4208/4393 [5:42:21<17:43, 5.75s/it] + Training...: 96% 4209/4393 [5:42:26<17:45, 5.79s/it] + Training...: 96% 4210/4393 [5:42:32<17:43, 5.81s/it] + Training...: 96% 4211/4393 [5:42:38<17:36, 5.81s/it] + Training...: 96% 4212/4393 [5:42:44<17:27, 5.79s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:19:50<17:31:17, 21025.84s/it] + Training...: 96% 4212/4393 [5:42:50<17:27, 5.79s/it] + Training...: 96% 4213/4393 [5:42:50<17:59, 6.00s/it] + Training...: 96% 4214/4393 [5:42:56<17:33, 5.89s/it] + Training...: 96% 4215/4393 [5:43:02<17:20, 5.85s/it] + Training...: 96% 4216/4393 [5:43:07<17:03, 5.78s/it] + Training...: 96% 4217/4393 [5:43:13<16:48, 5.73s/it] + Training...: 96% 4218/4393 [5:43:18<16:33, 5.68s/it] + Training...: 96% 4219/4393 [5:43:24<16:20, 5.64s/it] + Training...: 96% 4220/4393 [5:43:30<16:09, 5.61s/it] + Training...: 96% 4221/4393 [5:43:35<15:56, 5.56s/it] + Training...: 96% 4222/4393 [5:43:40<15:43, 5.52s/it] + Training...: 96% 4223/4393 [5:43:46<15:36, 5.51s/it] + Training...: 96% 4224/4393 [5:43:51<15:20, 5.45s/it] + Training...: 96% 4225/4393 [5:43:57<15:09, 5.42s/it] + Training...: 96% 4226/4393 [5:44:02<14:54, 5.35s/it] + Training...: 96% 4227/4393 [5:44:07<14:41, 5.31s/it] + Training...: 96% 4228/4393 [5:44:12<14:27, 5.26s/it] + Training...: 96% 4229/4393 [5:44:17<14:13, 5.20s/it] + Training...: 96% 4230/4393 [5:44:22<13:57, 5.14s/it] + Training...: 96% 4231/4393 [5:44:27<13:41, 5.07s/it] + Training...: 96% 4232/4393 [5:44:32<13:30, 5.04s/it] + Training...: 96% 4233/4393 [5:44:37<13:18, 4.99s/it] + Training...: 96% 4234/4393 [5:44:42<12:58, 4.90s/it] + Training...: 96% 4235/4393 [5:44:46<12:36, 4.79s/it] + Training...: 96% 4236/4393 [5:44:50<12:09, 4.65s/it] + Training...: 96% 4237/4393 [5:44:55<11:43, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:21:59<17:31:17, 21025.84s/it] + Training...: 96% 4237/4393 [5:44:59<11:43, 4.51s/it] + Training...: 96% 4238/4393 [5:44:59<11:43, 4.54s/it] + Training...: 96% 4239/4393 [5:45:03<11:05, 4.32s/it] + Training...: 97% 4240/4393 [5:45:07<10:26, 4.10s/it] + Training...: 97% 4241/4393 [5:45:10<09:48, 3.87s/it] + Training...: 97% 4242/4393 [5:45:13<09:12, 3.66s/it] + Training...: 97% 4243/4393 [5:45:16<08:37, 3.45s/it] + Training...: 97% 4244/4393 [5:45:19<07:59, 3.22s/it] + Training...: 97% 4245/4393 [5:45:21<07:19, 2.97s/it] + Training...: 97% 4246/4393 [5:45:23<06:40, 2.72s/it] + Training...: 97% 4247/4393 [5:45:25<05:58, 2.46s/it] + Training...: 97% 4248/4393 [5:45:27<05:18, 2.20s/it] + Training...: 97% 4249/4393 [5:45:28<04:38, 1.93s/it] + Training...: 97% 4250/4393 [5:45:29<03:58, 1.67s/it] + Training...: 97% 4251/4393 [5:45:35<07:09, 3.03s/it] + Training...: 97% 4252/4393 [5:45:42<09:22, 3.99s/it] + Training...: 97% 4253/4393 [5:45:48<10:50, 4.64s/it] + Training...: 97% 4254/4393 [5:45:54<11:43, 5.06s/it] + Training...: 97% 4255/4393 [5:46:00<12:19, 5.36s/it] + Training...: 97% 4256/4393 [5:46:06<12:38, 5.53s/it] + Training...: 97% 4257/4393 [5:46:12<12:50, 5.66s/it] + Training...: 97% 4258/4393 [5:46:18<12:53, 5.73s/it] + Training...: 97% 4259/4393 [5:46:23<12:53, 5.77s/it] + Training...: 97% 4260/4393 [5:46:29<12:48, 5.78s/it] + Training...: 97% 4261/4393 [5:46:35<12:46, 5.81s/it] + Training...: 97% 4262/4393 [5:46:41<12:41, 5.82s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:23:47<17:31:17, 21025.84s/it] + Training...: 97% 4262/4393 [5:46:48<12:41, 5.82s/it] + Training...: 97% 4263/4393 [5:46:48<13:07, 6.05s/it] + Training...: 97% 4264/4393 [5:46:53<12:47, 5.95s/it] + Training...: 97% 4265/4393 [5:46:59<12:30, 5.87s/it] + Training...: 97% 4266/4393 [5:47:05<12:19, 5.83s/it] + Training...: 97% 4267/4393 [5:47:10<12:11, 5.80s/it] + Training...: 97% 4268/4393 [5:47:16<11:58, 5.75s/it] + Training...: 97% 4269/4393 [5:47:22<11:44, 5.68s/it] + Training...: 97% 4270/4393 [5:47:27<11:28, 5.60s/it] + Training...: 97% 4271/4393 [5:47:32<11:15, 5.53s/it] + Training...: 97% 4272/4393 [5:47:38<11:02, 5.48s/it] + Training...: 97% 4273/4393 [5:47:43<10:53, 5.44s/it] + Training...: 97% 4274/4393 [5:47:48<10:42, 5.40s/it] + Training...: 97% 4275/4393 [5:47:54<10:30, 5.35s/it] + Training...: 97% 4276/4393 [5:47:59<10:19, 5.30s/it] + Training...: 97% 4277/4393 [5:48:04<10:08, 5.25s/it] + Training...: 97% 4278/4393 [5:48:09<09:58, 5.20s/it] + Training...: 97% 4279/4393 [5:48:14<09:48, 5.16s/it] + Training...: 97% 4280/4393 [5:48:19<09:35, 5.09s/it] + Training...: 97% 4281/4393 [5:48:24<09:25, 5.05s/it] + Training...: 97% 4282/4393 [5:48:29<09:11, 4.97s/it] + Training...: 97% 4283/4393 [5:48:33<08:59, 4.91s/it] + Training...: 98% 4284/4393 [5:48:38<08:45, 4.82s/it] + Training...: 98% 4285/4393 [5:48:43<08:32, 4.75s/it] + Training...: 98% 4286/4393 [5:48:47<08:17, 4.65s/it] + Training...: 98% 4287/4393 [5:48:51<08:04, 4.57s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:25:56<17:31:17, 21025.84s/it] + Training...: 98% 4287/4393 [5:48:56<08:04, 4.57s/it] + Training...: 98% 4288/4393 [5:48:56<08:07, 4.64s/it] + Training...: 98% 4289/4393 [5:49:00<07:41, 4.44s/it] + Training...: 98% 4290/4393 [5:49:04<07:13, 4.21s/it] + Training...: 98% 4291/4393 [5:49:07<06:45, 3.98s/it] + Training...: 98% 4292/4393 [5:49:11<06:18, 3.75s/it] + Training...: 98% 4293/4393 [5:49:14<05:51, 3.51s/it] + Training...: 98% 4294/4393 [5:49:16<05:23, 3.26s/it] + Training...: 98% 4295/4393 [5:49:19<04:53, 2.99s/it] + Training...: 98% 4296/4393 [5:49:21<04:25, 2.74s/it] + Training...: 98% 4297/4393 [5:49:23<03:56, 2.46s/it] + Training...: 98% 4298/4393 [5:49:24<03:29, 2.20s/it] + Training...: 98% 4299/4393 [5:49:25<03:01, 1.94s/it] + Training...: 98% 4300/4393 [5:49:27<02:35, 1.67s/it] + Training...: 98% 4301/4393 [5:49:33<04:39, 3.04s/it] + Training...: 98% 4302/4393 [5:49:39<06:04, 4.01s/it] + Training...: 98% 4303/4393 [5:49:45<06:57, 4.64s/it] + Training...: 98% 4304/4393 [5:49:51<07:31, 5.07s/it] + Training...: 98% 4305/4393 [5:49:57<07:53, 5.38s/it] + Training...: 98% 4306/4393 [5:50:03<08:05, 5.58s/it] + Training...: 98% 4307/4393 [5:50:09<08:09, 5.69s/it] + Training...: 98% 4308/4393 [5:50:15<08:09, 5.76s/it] + Training...: 98% 4309/4393 [5:50:21<08:08, 5.81s/it] + Training...: 98% 4310/4393 [5:50:27<08:02, 5.82s/it] + Training...: 98% 4311/4393 [5:50:33<07:59, 5.84s/it] + Training...: 98% 4312/4393 [5:50:39<07:59, 5.92s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:27:46<17:31:17, 21025.84s/it] + Training...: 98% 4312/4393 [5:50:46<07:59, 5.92s/it] + Training...: 98% 4313/4393 [5:50:46<08:12, 6.16s/it] + Training...: 98% 4314/4393 [5:50:51<07:56, 6.03s/it] + Training...: 98% 4315/4393 [5:50:57<07:43, 5.94s/it] + Training...: 98% 4316/4393 [5:51:03<07:29, 5.84s/it] + Training...: 98% 4317/4393 [5:51:08<07:17, 5.76s/it] + Training...: 98% 4318/4393 [5:51:14<07:07, 5.70s/it] + Training...: 98% 4319/4393 [5:51:19<06:57, 5.64s/it] + Training...: 98% 4320/4393 [5:51:25<06:47, 5.59s/it] + Training...: 98% 4321/4393 [5:51:30<06:38, 5.54s/it] + Training...: 98% 4322/4393 [5:51:36<06:30, 5.50s/it] + Training...: 98% 4323/4393 [5:51:41<06:23, 5.48s/it] + Training...: 98% 4324/4393 [5:51:47<06:15, 5.44s/it] + Training...: 98% 4325/4393 [5:51:52<06:07, 5.40s/it] + Training...: 98% 4326/4393 [5:51:57<05:59, 5.37s/it] + Training...: 98% 4327/4393 [5:52:02<05:52, 5.34s/it] + Training...: 99% 4328/4393 [5:52:08<05:47, 5.34s/it] + Training...: 99% 4329/4393 [5:52:13<05:39, 5.30s/it] + Training...: 99% 4330/4393 [5:52:18<05:27, 5.21s/it] + Training...: 99% 4331/4393 [5:52:23<05:17, 5.12s/it] + Training...: 99% 4332/4393 [5:52:28<05:06, 5.03s/it] + Training...: 99% 4333/4393 [5:52:32<04:55, 4.92s/it] + Training...: 99% 4334/4393 [5:52:37<04:45, 4.83s/it] + Training...: 99% 4335/4393 [5:52:42<04:35, 4.75s/it] + Training...: 99% 4336/4393 [5:52:46<04:24, 4.64s/it] + Training...: 99% 4337/4393 [5:52:50<04:12, 4.51s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:29:55<17:31:17, 21025.84s/it] + Training...: 99% 4337/4393 [5:52:55<04:12, 4.51s/it] + Training...: 99% 4338/4393 [5:52:55<04:11, 4.57s/it] + Training...: 99% 4339/4393 [5:52:59<03:53, 4.33s/it] + Training...: 99% 4340/4393 [5:53:02<03:35, 4.07s/it] + Training...: 99% 4341/4393 [5:53:05<03:19, 3.83s/it] + Training...: 99% 4342/4393 [5:53:08<03:02, 3.58s/it] + Training...: 99% 4343/4393 [5:53:11<02:47, 3.35s/it] + Training...: 99% 4344/4393 [5:53:14<02:32, 3.11s/it] + Training...: 99% 4345/4393 [5:53:16<02:16, 2.85s/it] + Training...: 99% 4346/4393 [5:53:18<02:02, 2.61s/it] + Training...: 99% 4347/4393 [5:53:20<01:48, 2.37s/it] + Training...: 99% 4348/4393 [5:53:21<01:35, 2.11s/it] + Training...: 99% 4349/4393 [5:53:23<01:21, 1.86s/it] + Training...: 99% 4350/4393 [5:53:24<01:09, 1.61s/it] + Training...: 99% 4351/4393 [5:53:30<02:07, 3.04s/it] + Training...: 99% 4352/4393 [5:53:36<02:44, 4.02s/it] + Training...: 99% 4353/4393 [5:53:43<03:07, 4.69s/it] + Training...: 99% 4354/4393 [5:53:49<03:20, 5.14s/it] + Training...: 99% 4355/4393 [5:53:55<03:30, 5.53s/it] + Training...: 99% 4356/4393 [5:54:01<03:31, 5.70s/it] + Training...: 99% 4357/4393 [5:54:07<03:28, 5.80s/it] + Training...: 99% 4358/4393 [5:54:13<03:24, 5.83s/it] + Training...: 99% 4359/4393 [5:54:19<03:20, 5.90s/it] + Training...: 99% 4360/4393 [5:54:25<03:14, 5.89s/it] + Training...: 99% 4361/4393 [5:54:31<03:08, 5.90s/it] + Training...: 99% 4362/4393 [5:54:37<03:03, 5.90s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:31:44<17:31:17, 21025.84s/it] + Training...: 99% 4362/4393 [5:54:44<03:03, 5.90s/it] + Training...: 99% 4363/4393 [5:54:44<03:04, 6.14s/it] + Training...: 99% 4364/4393 [5:54:49<02:53, 5.99s/it] + Training...: 99% 4365/4393 [5:54:55<02:44, 5.88s/it] + Training...: 99% 4366/4393 [5:55:00<02:35, 5.77s/it] + Training...: 99% 4367/4393 [5:55:06<02:28, 5.69s/it] + Training...: 99% 4368/4393 [5:55:11<02:20, 5.61s/it] + Training...: 99% 4369/4393 [5:55:17<02:13, 5.58s/it] + Training...: 99% 4370/4393 [5:55:23<02:09, 5.61s/it] + Training...: 99% 4371/4393 [5:55:28<02:02, 5.55s/it] + Training...: 100% 4372/4393 [5:55:33<01:54, 5.45s/it] + Training...: 100% 4373/4393 [5:55:38<01:47, 5.39s/it] + Training...: 100% 4374/4393 [5:55:44<01:40, 5.31s/it] + Training...: 100% 4375/4393 [5:55:49<01:34, 5.25s/it] + Training...: 100% 4376/4393 [5:55:54<01:27, 5.17s/it] + Training...: 100% 4377/4393 [5:55:59<01:21, 5.11s/it] + Training...: 100% 4378/4393 [5:56:03<01:15, 5.02s/it] + Training...: 100% 4379/4393 [5:56:08<01:08, 4.92s/it] + Training...: 100% 4380/4393 [5:56:13<01:02, 4.82s/it] + Training...: 100% 4381/4393 [5:56:17<00:56, 4.72s/it] + Training...: 100% 4382/4393 [5:56:21<00:50, 4.59s/it] + Training...: 100% 4383/4393 [5:56:26<00:44, 4.44s/it] + Training...: 100% 4384/4393 [5:56:29<00:38, 4.27s/it] + Training...: 100% 4385/4393 [5:56:33<00:32, 4.08s/it] + Training...: 100% 4386/4393 [5:56:36<00:27, 3.86s/it] + Training...: 100% 4387/4393 [5:56:39<00:21, 3.59s/it] +  Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 75% 9/12 [58:33:42<17:31:17, 21025.84s/it] + Training...: 100% 4387/4393 [5:56:42<00:21, 3.59s/it] + Training...: 100% 4388/4393 [5:56:42<00:17, 3.44s/it] + Training...: 100% 4389/4393 [5:56:45<00:12, 3.10s/it] + Training...: 100% 4390/4393 [5:56:47<00:08, 2.77s/it] + Training...: 100% 4391/4393 [5:56:48<00:04, 2.45s/it] + Training...: 100% 4392/4393 [5:56:50<00:02, 2.13s/it] + Training...: 100% 4393/4393 [5:56:51<00:00, 1.84s/it] Training...: 100% 4393/4393 [5:56:51<00:00, 4.87s/it] + Step... (40000/50000 | Eval Loss: 0.9527401328086853 | Eval wer: 0.04187346053453917 | Eval cer: 0.02872188479352137 |): 83% 10/12 [58:33:51<11:44:51, 21145.58s/it] +Step... (43400 | Loss: 0.006220472510904074, Learning Rate: 1.3335352377907839e-05, Gradient Norm: 0.15546762943267822) +Step... (43425 | Loss: 0.006581796798855066, Learning Rate: 1.3284849956107792e-05, Gradient Norm: 0.34990987181663513) +Step... (43450 | Loss: 0.011822747066617012, Learning Rate: 1.3234341167844832e-05, Gradient Norm: 0.2574056386947632) +Step... (43475 | Loss: 0.00934094749391079, Learning Rate: 1.3183837836550083e-05, Gradient Norm: 0.3183192014694214) +Step... (43500 | Loss: 0.005102855619043112, Learning Rate: 1.3133334505255334e-05, Gradient Norm: 0.14597401022911072) +Step... (43525 | Loss: 0.008145734667778015, Learning Rate: 1.3082825716992375e-05, Gradient Norm: 0.19706687331199646) +Step... (43550 | Loss: 0.009935390204191208, Learning Rate: 1.3032322385697626e-05, Gradient Norm: 0.26017606258392334) +Step... (43575 | Loss: 0.007766792085021734, Learning Rate: 1.2981819963897578e-05, Gradient Norm: 0.2769607603549957) +Step... (43600 | Loss: 0.010773919522762299, Learning Rate: 1.2931311175634619e-05, Gradient Norm: 0.23994655907154083) +Step... (43625 | Loss: 0.0029236481059342623, Learning Rate: 1.288080784433987e-05, Gradient Norm: 0.1163998618721962) +Step... (43650 | Loss: 0.013007597997784615, Learning Rate: 1.2830304513045121e-05, Gradient Norm: 0.2571600377559662) +Step... (43675 | Loss: 0.006016268394887447, Learning Rate: 1.2779795724782161e-05, Gradient Norm: 0.2098936140537262) +Step... (43700 | Loss: 0.00377640244551003, Learning Rate: 1.2729292393487412e-05, Gradient Norm: 0.0722104161977768) +Step... (43725 | Loss: 0.005282220430672169, Learning Rate: 1.2678789971687365e-05, Gradient Norm: 0.19065354764461517) +Step... (43750 | Loss: 0.005376732908189297, Learning Rate: 1.2628281183424406e-05, Gradient Norm: 0.16388669610023499) +Step... (43775 | Loss: 0.008202340453863144, Learning Rate: 1.2577777852129657e-05, Gradient Norm: 0.1876869797706604) +Step... (43800 | Loss: 0.004423276986926794, Learning Rate: 1.2527274520834908e-05, Gradient Norm: 0.14415715634822845) +Step... (43825 | Loss: 0.0041627101600170135, Learning Rate: 1.2476765732571948e-05, Gradient Norm: 0.18802481889724731) +Step... (43850 | Loss: 0.012115256860852242, Learning Rate: 1.24262624012772e-05, Gradient Norm: 0.35901159048080444) +Step... (43875 | Loss: 0.005729792173951864, Learning Rate: 1.2375759979477152e-05, Gradient Norm: 0.3747013211250305) +Step... (43900 | Loss: 0.005576059687882662, Learning Rate: 1.2325251191214193e-05, Gradient Norm: 0.14393489062786102) +Step... (43925 | Loss: 0.005378743167966604, Learning Rate: 1.2274747859919444e-05, Gradient Norm: 0.26317912340164185) + Training...: 0% 0/4393 [00:00