Training in progress, step 2500, checkpoint

checkpoint-2500/added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<ctc_blank>": 80,
+  "<mask>": 79
+}

checkpoint-2500/config.json

@@ -0,0 +1,92 @@
+{
+  "_name_or_path": "microsoft/speecht5_tts",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "SpeechT5ForTextToSpeech"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 0,
+  "conv_bias": false,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "decoder_attention_heads": 12,
+  "decoder_ffn_dim": 3072,
+  "decoder_layerdrop": 0.1,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 2,
+  "encoder_attention_heads": 12,
+  "encoder_ffn_dim": 3072,
+  "encoder_layerdrop": 0.1,
+  "encoder_layers": 12,
+  "encoder_max_relative_position": 160,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.0,
+  "guided_attention_loss_num_heads": 2,
+  "guided_attention_loss_scale": 10.0,
+  "guided_attention_loss_sigma": 0.4,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "is_encoder_decoder": true,
+  "layer_norm_eps": 1e-05,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "max_length": 1876,
+  "max_speech_positions": 1876,
+  "max_text_positions": 600,
+  "model_type": "speecht5",
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_mel_bins": 80,
+  "pad_token_id": 1,
+  "positional_dropout": 0.1,
+  "reduction_factor": 2,
+  "scale_embedding": false,
+  "speaker_embedding_dim": 512,
+  "speech_decoder_postnet_dropout": 0.5,
+  "speech_decoder_postnet_kernel": 5,
+  "speech_decoder_postnet_layers": 5,
+  "speech_decoder_postnet_units": 256,
+  "speech_decoder_prenet_dropout": 0.5,
+  "speech_decoder_prenet_layers": 2,
+  "speech_decoder_prenet_units": 256,
+  "torch_dtype": "float32",
+  "transformers_version": "4.41.0.dev0",
+  "use_cache": false,
+  "use_guided_attention_loss": true,
+  "vocab_size": 81
+}

checkpoint-2500/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "eos_token_id": 2,
+  "max_length": 1876,
+  "pad_token_id": 1,
+  "transformers_version": "4.41.0.dev0"
+}

checkpoint-2500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e45d0ba53fae0007185fa228c23f8efb2cc1a670a55e18b5c6ee49b0235f76d0
+size 577789320

checkpoint-2500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4223bcfb72cbb883f3580c91748d43a99d5f7e40c74ba335c57478020822fb34
+size 1155772233

checkpoint-2500/preprocessor_config.json

@@ -0,0 +1,19 @@
+{
+  "do_normalize": false,
+  "feature_extractor_type": "SpeechT5FeatureExtractor",
+  "feature_size": 1,
+  "fmax": 7600,
+  "fmin": 80,
+  "frame_signal_scale": 1.0,
+  "hop_length": 16,
+  "mel_floor": 1e-10,
+  "num_mel_bins": 80,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "SpeechT5Processor",
+  "reduction_factor": 2,
+  "return_attention_mask": true,
+  "sampling_rate": 16000,
+  "win_function": "hann_window",
+  "win_length": 64
+}

checkpoint-2500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c399ba4fe58f486c2ad08014b04a703682ee373818118e2b9e4d9f8bd74ec312
+size 14244

checkpoint-2500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a6e1d45386e11e147f95efa198dfdb563b15409648099710720a92d2030eff3
+size 1064

checkpoint-2500/special_tokens_map.json

@@ -0,0 +1,13 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-2500/spm_char.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
+size 238473

checkpoint-2500/tokenizer_config.json

@@ -0,0 +1,63 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<ctc_blank>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 600,
+  "normalize": false,
+  "pad_token": "<pad>",
+  "processor_class": "SpeechT5Processor",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "SpeechT5Tokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-2500/trainer_state.json

@@ -0,0 +1,423 @@
+{
+  "best_metric": 0.41808074712753296,
+  "best_model_checkpoint": "mikhail-panzo/fil_b32_le4_s4000/checkpoint-2500",
+  "epoch": 55.55555555555556,
+  "eval_steps": 500,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 1.5478250980377197,
+      "learning_rate": 2.5e-06,
+      "loss": 0.7933,
+      "step": 50
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 2.5992791652679443,
+      "learning_rate": 5e-06,
+      "loss": 0.7126,
+      "step": 100
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 4.512153625488281,
+      "learning_rate": 7.45e-06,
+      "loss": 0.6643,
+      "step": 150
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 5.529618740081787,
+      "learning_rate": 9.950000000000001e-06,
+      "loss": 0.6161,
+      "step": 200
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 1.547269344329834,
+      "learning_rate": 1.2450000000000001e-05,
+      "loss": 0.5572,
+      "step": 250
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.7310105562210083,
+      "learning_rate": 1.4950000000000001e-05,
+      "loss": 0.5307,
+      "step": 300
+    },
+    {
+      "epoch": 7.777777777777778,
+      "grad_norm": 4.040494441986084,
+      "learning_rate": 1.745e-05,
+      "loss": 0.5259,
+      "step": 350
+    },
+    {
+      "epoch": 8.88888888888889,
+      "grad_norm": 1.6819641590118408,
+      "learning_rate": 1.995e-05,
+      "loss": 0.4981,
+      "step": 400
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.428649663925171,
+      "learning_rate": 2.245e-05,
+      "loss": 0.5008,
+      "step": 450
+    },
+    {
+      "epoch": 11.11111111111111,
+      "grad_norm": 2.2339069843292236,
+      "learning_rate": 2.495e-05,
+      "loss": 0.4923,
+      "step": 500
+    },
+    {
+      "epoch": 11.11111111111111,
+      "eval_loss": 0.445181667804718,
+      "eval_runtime": 8.1426,
+      "eval_samples_per_second": 19.527,
+      "eval_steps_per_second": 2.456,
+      "step": 500
+    },
+    {
+      "epoch": 12.222222222222221,
+      "grad_norm": 3.6943907737731934,
+      "learning_rate": 2.7450000000000003e-05,
+      "loss": 0.4999,
+      "step": 550
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 2.8963587284088135,
+      "learning_rate": 2.995e-05,
+      "loss": 0.4723,
+      "step": 600
+    },
+    {
+      "epoch": 14.444444444444445,
+      "grad_norm": 2.157615900039673,
+      "learning_rate": 3.245e-05,
+      "loss": 0.4771,
+      "step": 650
+    },
+    {
+      "epoch": 15.555555555555555,
+      "grad_norm": 4.2828450202941895,
+      "learning_rate": 3.495e-05,
+      "loss": 0.4656,
+      "step": 700
+    },
+    {
+      "epoch": 16.666666666666668,
+      "grad_norm": 2.30914044380188,
+      "learning_rate": 3.745e-05,
+      "loss": 0.4681,
+      "step": 750
+    },
+    {
+      "epoch": 17.77777777777778,
+      "grad_norm": 1.7412217855453491,
+      "learning_rate": 3.995e-05,
+      "loss": 0.4639,
+      "step": 800
+    },
+    {
+      "epoch": 18.88888888888889,
+      "grad_norm": 2.457113742828369,
+      "learning_rate": 4.245e-05,
+      "loss": 0.4691,
+      "step": 850
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 2.960203170776367,
+      "learning_rate": 4.495e-05,
+      "loss": 0.4569,
+      "step": 900
+    },
+    {
+      "epoch": 21.11111111111111,
+      "grad_norm": 3.9787750244140625,
+      "learning_rate": 4.745e-05,
+      "loss": 0.4547,
+      "step": 950
+    },
+    {
+      "epoch": 22.22222222222222,
+      "grad_norm": 1.6046252250671387,
+      "learning_rate": 4.995e-05,
+      "loss": 0.4488,
+      "step": 1000
+    },
+    {
+      "epoch": 22.22222222222222,
+      "eval_loss": 0.42208045721054077,
+      "eval_runtime": 7.5792,
+      "eval_samples_per_second": 20.978,
+      "eval_steps_per_second": 2.639,
+      "step": 1000
+    },
+    {
+      "epoch": 23.333333333333332,
+      "grad_norm": 2.2585225105285645,
+      "learning_rate": 5.245e-05,
+      "loss": 0.4537,
+      "step": 1050
+    },
+    {
+      "epoch": 24.444444444444443,
+      "grad_norm": 2.7472567558288574,
+      "learning_rate": 5.495e-05,
+      "loss": 0.4438,
+      "step": 1100
+    },
+    {
+      "epoch": 25.555555555555557,
+      "grad_norm": 2.1399290561676025,
+      "learning_rate": 5.745e-05,
+      "loss": 0.4474,
+      "step": 1150
+    },
+    {
+      "epoch": 26.666666666666668,
+      "grad_norm": 2.671912670135498,
+      "learning_rate": 5.995000000000001e-05,
+      "loss": 0.4488,
+      "step": 1200
+    },
+    {
+      "epoch": 27.77777777777778,
+      "grad_norm": 4.835938453674316,
+      "learning_rate": 6.245000000000001e-05,
+      "loss": 0.4472,
+      "step": 1250
+    },
+    {
+      "epoch": 28.88888888888889,
+      "grad_norm": 5.086827754974365,
+      "learning_rate": 6.494999999999999e-05,
+      "loss": 0.4526,
+      "step": 1300
+    },
+    {
+      "epoch": 30.0,
+      "grad_norm": 2.293753147125244,
+      "learning_rate": 6.745e-05,
+      "loss": 0.4448,
+      "step": 1350
+    },
+    {
+      "epoch": 31.11111111111111,
+      "grad_norm": 3.609741449356079,
+      "learning_rate": 6.995e-05,
+      "loss": 0.438,
+      "step": 1400
+    },
+    {
+      "epoch": 32.22222222222222,
+      "grad_norm": 1.580810546875,
+      "learning_rate": 7.245000000000001e-05,
+      "loss": 0.4446,
+      "step": 1450
+    },
+    {
+      "epoch": 33.333333333333336,
+      "grad_norm": 2.7843821048736572,
+      "learning_rate": 7.495e-05,
+      "loss": 0.4413,
+      "step": 1500
+    },
+    {
+      "epoch": 33.333333333333336,
+      "eval_loss": 0.41924577951431274,
+      "eval_runtime": 7.8331,
+      "eval_samples_per_second": 20.298,
+      "eval_steps_per_second": 2.553,
+      "step": 1500
+    },
+    {
+      "epoch": 34.44444444444444,
+      "grad_norm": 5.7151875495910645,
+      "learning_rate": 7.745e-05,
+      "loss": 0.4369,
+      "step": 1550
+    },
+    {
+      "epoch": 35.55555555555556,
+      "grad_norm": 2.420909881591797,
+      "learning_rate": 7.995e-05,
+      "loss": 0.4491,
+      "step": 1600
+    },
+    {
+      "epoch": 36.666666666666664,
+      "grad_norm": 3.388465642929077,
+      "learning_rate": 8.245e-05,
+      "loss": 0.4381,
+      "step": 1650
+    },
+    {
+      "epoch": 37.77777777777778,
+      "grad_norm": 1.8416664600372314,
+      "learning_rate": 8.495e-05,
+      "loss": 0.4332,
+      "step": 1700
+    },
+    {
+      "epoch": 38.888888888888886,
+      "grad_norm": 4.991604328155518,
+      "learning_rate": 8.745000000000001e-05,
+      "loss": 0.4363,
+      "step": 1750
+    },
+    {
+      "epoch": 40.0,
+      "grad_norm": 1.6002848148345947,
+      "learning_rate": 8.995e-05,
+      "loss": 0.4296,
+      "step": 1800
+    },
+    {
+      "epoch": 41.111111111111114,
+      "grad_norm": 2.4206886291503906,
+      "learning_rate": 9.245e-05,
+      "loss": 0.4273,
+      "step": 1850
+    },
+    {
+      "epoch": 42.22222222222222,
+      "grad_norm": 3.848668098449707,
+      "learning_rate": 9.495e-05,
+      "loss": 0.4244,
+      "step": 1900
+    },
+    {
+      "epoch": 43.333333333333336,
+      "grad_norm": 3.4843313694000244,
+      "learning_rate": 9.745000000000001e-05,
+      "loss": 0.4213,
+      "step": 1950
+    },
+    {
+      "epoch": 44.44444444444444,
+      "grad_norm": 7.2384772300720215,
+      "learning_rate": 9.995e-05,
+      "loss": 0.4318,
+      "step": 2000
+    },
+    {
+      "epoch": 44.44444444444444,
+      "eval_loss": 0.42309698462486267,
+      "eval_runtime": 7.2889,
+      "eval_samples_per_second": 21.814,
+      "eval_steps_per_second": 2.744,
+      "step": 2000
+    },
+    {
+      "epoch": 45.55555555555556,
+      "grad_norm": 1.936278223991394,
+      "learning_rate": 9.755000000000001e-05,
+      "loss": 0.4279,
+      "step": 2050
+    },
+    {
+      "epoch": 46.666666666666664,
+      "grad_norm": 6.694615840911865,
+      "learning_rate": 9.505e-05,
+      "loss": 0.4326,
+      "step": 2100
+    },
+    {
+      "epoch": 47.77777777777778,
+      "grad_norm": 3.927391529083252,
+      "learning_rate": 9.260000000000001e-05,
+      "loss": 0.434,
+      "step": 2150
+    },
+    {
+      "epoch": 48.888888888888886,
+      "grad_norm": 2.543116569519043,
+      "learning_rate": 9.010000000000001e-05,
+      "loss": 0.4338,
+      "step": 2200
+    },
+    {
+      "epoch": 50.0,
+      "grad_norm": 3.7880661487579346,
+      "learning_rate": 8.76e-05,
+      "loss": 0.4239,
+      "step": 2250
+    },
+    {
+      "epoch": 51.111111111111114,
+      "grad_norm": 6.448720455169678,
+      "learning_rate": 8.510000000000001e-05,
+      "loss": 0.423,
+      "step": 2300
+    },
+    {
+      "epoch": 52.22222222222222,
+      "grad_norm": 2.7751946449279785,
+      "learning_rate": 8.26e-05,
+      "loss": 0.4182,
+      "step": 2350
+    },
+    {
+      "epoch": 53.333333333333336,
+      "grad_norm": 3.261136054992676,
+      "learning_rate": 8.010000000000001e-05,
+      "loss": 0.4133,
+      "step": 2400
+    },
+    {
+      "epoch": 54.44444444444444,
+      "grad_norm": 3.6887683868408203,
+      "learning_rate": 7.76e-05,
+      "loss": 0.4096,
+      "step": 2450
+    },
+    {
+      "epoch": 55.55555555555556,
+      "grad_norm": 2.9640533924102783,
+      "learning_rate": 7.510000000000001e-05,
+      "loss": 0.4155,
+      "step": 2500
+    },
+    {
+      "epoch": 55.55555555555556,
+      "eval_loss": 0.41808074712753296,
+      "eval_runtime": 7.7845,
+      "eval_samples_per_second": 20.425,
+      "eval_steps_per_second": 2.569,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 4000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 89,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.722388964755536e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7395c0a6b369ee3197719256b2f1828d445ab49e677f945a930ff4142cbbb1ee
+size 5304