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:d43efe3a5c72d30831f85e9691fd70827868b886ad4736d4ab4c746a35e97963
+size 577789320

checkpoint-2500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:019669c49b4ba02d6de3469a2d6a059ee4cc02cd6a0357b092a24261cd7feedb
+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 sha256:6897fa3d8d9743513473619f7423066e927c20bf6f50c5777bff0318ac4e29ec
+size 14244

checkpoint-2500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a70fbbaadfec6c2080573fea9ba9b136cf835b11da700f1c3ccce349e3a02ff8
+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": 0.395076185464859,
+  "best_model_checkpoint": "mikhail-panzo/ceb_b128_le4_s4000/checkpoint-1500",
+  "epoch": 198.01980198019803,
+  "eval_steps": 500,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 2.952680826187134,
+      "learning_rate": 2.5e-06,
+      "loss": 0.7707,
+      "step": 50
+    },
+    {
+      "epoch": 7.920792079207921,
+      "grad_norm": 1.126000165939331,
+      "learning_rate": 5e-06,
+      "loss": 0.6934,
+      "step": 100
+    },
+    {
+      "epoch": 11.881188118811881,
+      "grad_norm": 2.129329204559326,
+      "learning_rate": 7.5e-06,
+      "loss": 0.5982,
+      "step": 150
+    },
+    {
+      "epoch": 15.841584158415841,
+      "grad_norm": 1.1155015230178833,
+      "learning_rate": 1e-05,
+      "loss": 0.5093,
+      "step": 200
+    },
+    {
+      "epoch": 19.801980198019802,
+      "grad_norm": 1.3643360137939453,
+      "learning_rate": 1.25e-05,
+      "loss": 0.4912,
+      "step": 250
+    },
+    {
+      "epoch": 23.762376237623762,
+      "grad_norm": 0.9774149060249329,
+      "learning_rate": 1.5e-05,
+      "loss": 0.4764,
+      "step": 300
+    },
+    {
+      "epoch": 27.722772277227723,
+      "grad_norm": 0.8793525695800781,
+      "learning_rate": 1.75e-05,
+      "loss": 0.4648,
+      "step": 350
+    },
+    {
+      "epoch": 31.683168316831683,
+      "grad_norm": 0.7882798910140991,
+      "learning_rate": 2e-05,
+      "loss": 0.459,
+      "step": 400
+    },
+    {
+      "epoch": 35.64356435643565,
+      "grad_norm": 1.2033296823501587,
+      "learning_rate": 2.25e-05,
+      "loss": 0.4519,
+      "step": 450
+    },
+    {
+      "epoch": 39.603960396039604,
+      "grad_norm": 0.8620955348014832,
+      "learning_rate": 2.5e-05,
+      "loss": 0.4433,
+      "step": 500
+    },
+    {
+      "epoch": 39.603960396039604,
+      "eval_loss": 0.4056467115879059,
+      "eval_runtime": 7.4199,
+      "eval_samples_per_second": 24.259,
+      "eval_steps_per_second": 3.1,
+      "step": 500
+    },
+    {
+      "epoch": 43.56435643564357,
+      "grad_norm": 1.266213297843933,
+      "learning_rate": 2.7500000000000004e-05,
+      "loss": 0.444,
+      "step": 550
+    },
+    {
+      "epoch": 47.524752475247524,
+      "grad_norm": 1.174134373664856,
+      "learning_rate": 3e-05,
+      "loss": 0.437,
+      "step": 600
+    },
+    {
+      "epoch": 51.48514851485149,
+      "grad_norm": 0.8188551664352417,
+      "learning_rate": 3.2500000000000004e-05,
+      "loss": 0.4318,
+      "step": 650
+    },
+    {
+      "epoch": 55.445544554455445,
+      "grad_norm": 1.3083854913711548,
+      "learning_rate": 3.5e-05,
+      "loss": 0.4297,
+      "step": 700
+    },
+    {
+      "epoch": 59.40594059405941,
+      "grad_norm": 0.8094576001167297,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.4238,
+      "step": 750
+    },
+    {
+      "epoch": 63.366336633663366,
+      "grad_norm": 1.2506986856460571,
+      "learning_rate": 4e-05,
+      "loss": 0.4233,
+      "step": 800
+    },
+    {
+      "epoch": 67.32673267326733,
+      "grad_norm": 1.7902193069458008,
+      "learning_rate": 4.25e-05,
+      "loss": 0.4208,
+      "step": 850
+    },
+    {
+      "epoch": 71.2871287128713,
+      "grad_norm": 2.209643602371216,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4195,
+      "step": 900
+    },
+    {
+      "epoch": 75.24752475247524,
+      "grad_norm": 2.256385087966919,
+      "learning_rate": 4.75e-05,
+      "loss": 0.4202,
+      "step": 950
+    },
+    {
+      "epoch": 79.20792079207921,
+      "grad_norm": 0.9167351126670837,
+      "learning_rate": 5e-05,
+      "loss": 0.4142,
+      "step": 1000
+    },
+    {
+      "epoch": 79.20792079207921,
+      "eval_loss": 0.39529716968536377,
+      "eval_runtime": 6.6198,
+      "eval_samples_per_second": 27.191,
+      "eval_steps_per_second": 3.474,
+      "step": 1000
+    },
+    {
+      "epoch": 83.16831683168317,
+      "grad_norm": 0.9256124496459961,
+      "learning_rate": 5.25e-05,
+      "loss": 0.4137,
+      "step": 1050
+    },
+    {
+      "epoch": 87.12871287128714,
+      "grad_norm": 1.9655492305755615,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 0.4132,
+      "step": 1100
+    },
+    {
+      "epoch": 91.08910891089108,
+      "grad_norm": 1.08187997341156,
+      "learning_rate": 5.7499999999999995e-05,
+      "loss": 0.4078,
+      "step": 1150
+    },
+    {
+      "epoch": 95.04950495049505,
+      "grad_norm": 1.8732513189315796,
+      "learning_rate": 6e-05,
+      "loss": 0.4061,
+      "step": 1200
+    },
+    {
+      "epoch": 99.00990099009901,
+      "grad_norm": 1.3401920795440674,
+      "learning_rate": 6.25e-05,
+      "loss": 0.4054,
+      "step": 1250
+    },
+    {
+      "epoch": 102.97029702970298,
+      "grad_norm": 0.8879286646842957,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 0.4046,
+      "step": 1300
+    },
+    {
+      "epoch": 106.93069306930693,
+      "grad_norm": 1.2213553190231323,
+      "learning_rate": 6.750000000000001e-05,
+      "loss": 0.3999,
+      "step": 1350
+    },
+    {
+      "epoch": 110.89108910891089,
+      "grad_norm": 2.05886173248291,
+      "learning_rate": 7e-05,
+      "loss": 0.3983,
+      "step": 1400
+    },
+    {
+      "epoch": 114.85148514851485,
+      "grad_norm": 1.10814368724823,
+      "learning_rate": 7.25e-05,
+      "loss": 0.3964,
+      "step": 1450
+    },
+    {
+      "epoch": 118.81188118811882,
+      "grad_norm": 1.0220105648040771,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.3972,
+      "step": 1500
+    },
+    {
+      "epoch": 118.81188118811882,
+      "eval_loss": 0.395076185464859,
+      "eval_runtime": 7.1134,
+      "eval_samples_per_second": 25.304,
+      "eval_steps_per_second": 3.233,
+      "step": 1500
+    },
+    {
+      "epoch": 122.77227722772277,
+      "grad_norm": 1.1671476364135742,
+      "learning_rate": 7.75e-05,
+      "loss": 0.3909,
+      "step": 1550
+    },
+    {
+      "epoch": 126.73267326732673,
+      "grad_norm": 0.8167665600776672,
+      "learning_rate": 8e-05,
+      "loss": 0.3913,
+      "step": 1600
+    },
+    {
+      "epoch": 130.69306930693068,
+      "grad_norm": 1.9696727991104126,
+      "learning_rate": 8.25e-05,
+      "loss": 0.3888,
+      "step": 1650
+    },
+    {
+      "epoch": 134.65346534653466,
+      "grad_norm": 1.1445192098617554,
+      "learning_rate": 8.5e-05,
+      "loss": 0.3885,
+      "step": 1700
+    },
+    {
+      "epoch": 138.6138613861386,
+      "grad_norm": 1.908463954925537,
+      "learning_rate": 8.75e-05,
+      "loss": 0.3888,
+      "step": 1750
+    },
+    {
+      "epoch": 142.5742574257426,
+      "grad_norm": 1.6358473300933838,
+      "learning_rate": 9e-05,
+      "loss": 0.3878,
+      "step": 1800
+    },
+    {
+      "epoch": 146.53465346534654,
+      "grad_norm": 1.7407046556472778,
+      "learning_rate": 9.250000000000001e-05,
+      "loss": 0.3868,
+      "step": 1850
+    },
+    {
+      "epoch": 150.4950495049505,
+      "grad_norm": 3.984604597091675,
+      "learning_rate": 9.5e-05,
+      "loss": 0.3877,
+      "step": 1900
+    },
+    {
+      "epoch": 154.45544554455446,
+      "grad_norm": 1.5218504667282104,
+      "learning_rate": 9.75e-05,
+      "loss": 0.3828,
+      "step": 1950
+    },
+    {
+      "epoch": 158.41584158415841,
+      "grad_norm": 1.2710033655166626,
+      "learning_rate": 0.0001,
+      "loss": 0.3806,
+      "step": 2000
+    },
+    {
+      "epoch": 158.41584158415841,
+      "eval_loss": 0.39701494574546814,
+      "eval_runtime": 7.067,
+      "eval_samples_per_second": 25.47,
+      "eval_steps_per_second": 3.255,
+      "step": 2000
+    },
+    {
+      "epoch": 162.37623762376236,
+      "grad_norm": 1.905275583267212,
+      "learning_rate": 9.75e-05,
+      "loss": 0.3817,
+      "step": 2050
+    },
+    {
+      "epoch": 166.33663366336634,
+      "grad_norm": 2.0633037090301514,
+      "learning_rate": 9.5e-05,
+      "loss": 0.3805,
+      "step": 2100
+    },
+    {
+      "epoch": 170.2970297029703,
+      "grad_norm": 1.110232949256897,
+      "learning_rate": 9.250000000000001e-05,
+      "loss": 0.378,
+      "step": 2150
+    },
+    {
+      "epoch": 174.25742574257427,
+      "grad_norm": 2.0630099773406982,
+      "learning_rate": 9e-05,
+      "loss": 0.3742,
+      "step": 2200
+    },
+    {
+      "epoch": 178.21782178217822,
+      "grad_norm": 1.2983876466751099,
+      "learning_rate": 8.75e-05,
+      "loss": 0.3745,
+      "step": 2250
+    },
+    {
+      "epoch": 182.17821782178217,
+      "grad_norm": 0.6691053509712219,
+      "learning_rate": 8.5e-05,
+      "loss": 0.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 186.13861386138615,
+      "grad_norm": 1.2052526473999023,
+      "learning_rate": 8.25e-05,
+      "loss": 0.3719,
+      "step": 2350
+    },
+    {
+      "epoch": 190.0990099009901,
+      "grad_norm": 1.1967781782150269,
+      "learning_rate": 8e-05,
+      "loss": 0.3711,
+      "step": 2400
+    },
+    {
+      "epoch": 194.05940594059405,
+      "grad_norm": 0.8571879267692566,
+      "learning_rate": 7.75e-05,
+      "loss": 0.3715,
+      "step": 2450
+    },
+    {
+      "epoch": 198.01980198019803,
+      "grad_norm": 1.2208068370819092,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.3653,
+      "step": 2500
+    },
+    {
+      "epoch": 198.01980198019803,
+      "eval_loss": 0.3959733843803406,
+      "eval_runtime": 6.5933,
+      "eval_samples_per_second": 27.3,
+      "eval_steps_per_second": 3.488,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 4000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 334,
+  "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": 5.40291039934441e+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 sha256:1110c837d9cde38cbbbd08ff84eb79516ba15786ac829701460d74dc426431bb
+size 5304