Added acoustic model, language model

config.json

@@ -0,0 +1,117 @@
+{
+  "_name_or_path": "train/run_3/2022-03-31-10-42-03/checkpoint-6480",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "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
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "freeze_feat_extract_train": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "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_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
+  "num_adapter_layers": 3,
+  "num_attention_heads": 12,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 12,
+  "num_negatives": 100,
+  "output_hidden_size": 768,
+  "pad_token_id": 36,
+  "proj_codevector_dim": 256,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 39,
+  "xvector_output_dim": 512
+}

language_model/cv8be_5gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc3fa853d83b07d4de0b053d9bd6e2ac5154696e85d49fe3e94ed5b0102689a9
+size 155935664

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e49f65ffb62cb91604fa591c4794b3b48d760ec3a65fe477962de46c8c70c206
+size 721728273

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

processor/preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

processor/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}

processor/tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

processor/vocab.json

@@ -0,0 +1 @@
+{"'": 1, "i": 2, "а": 3, "б": 4, "в": 5, "г": 6, "д": 7, "е": 8, "ж": 9, "з": 10, "й": 11, "к": 12, "л": 13, "м": 14, "н": 15, "о": 16, "п": 17, "р": 18, "с": 19, "т": 20, "у": 21, "ф": 22, "х": 23, "ц": 24, "ч": 25, "ш": 26, "ы": 27, "ь": 28, "э": 29, "ю": 30, "я": 31, "ё": 32, "і": 33, "ў": 34, "|": 0, "[UNK]": 35, "[PAD]": 36}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b55c0aa0c1647ce563973ffa030dfacede44a21bae775b2bb4388e8e9a166f4f
+size 377693079

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b1697980a93af111ba6e84c6af7b2de4b85bc2bd1d3b82e487bf67b06745a81
+size 16607

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42c33f49f1b2c2aa4dfe87afa721ef222988d59691fcded6f07cef251e796c16
+size 559

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbf58a653951345865b70bd73aae4c04340c9cc086cd1f18d3a6751b20b1c7b3
+size 623

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_metric": 0.17608685300699967,
+  "best_model_checkpoint": "train/2022-03-31-21-50-25/checkpoint-12960",
+  "epoch": 1.9789280806229959,
+  "global_step": 12960,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 2e-05,
+      "loss": 0.1209,
+      "step": 1080
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 0.1341254562139511,
+      "eval_runtime": 468.0255,
+      "eval_samples_per_second": 33.765,
+      "eval_steps_per_second": 0.705,
+      "eval_wer": 0.19390346491436916,
+      "step": 1080
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8202695956065902e-05,
+      "loss": 0.1034,
+      "step": 2160
+    },
+    {
+      "epoch": 0.33,
+      "eval_loss": 0.13459239900112152,
+      "eval_runtime": 466.4831,
+      "eval_samples_per_second": 33.877,
+      "eval_steps_per_second": 0.707,
+      "eval_wer": 0.19213370791867054,
+      "step": 2160
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6405391912131803e-05,
+      "loss": 0.102,
+      "step": 3240
+    },
+    {
+      "epoch": 0.49,
+      "eval_loss": 0.13314320147037506,
+      "eval_runtime": 464.4247,
+      "eval_samples_per_second": 34.027,
+      "eval_steps_per_second": 0.711,
+      "eval_wer": 0.1909591606748885,
+      "step": 3240
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.4609752038608755e-05,
+      "loss": 0.1016,
+      "step": 4320
+    },
+    {
+      "epoch": 0.66,
+      "eval_loss": 0.13013887405395508,
+      "eval_runtime": 465.4732,
+      "eval_samples_per_second": 33.95,
+      "eval_steps_per_second": 0.709,
+      "eval_wer": 0.1890465533387299,
+      "step": 4320
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.2812447994674656e-05,
+      "loss": 0.109,
+      "step": 5400
+    },
+    {
+      "epoch": 0.82,
+      "eval_loss": 0.12718452513217926,
+      "eval_runtime": 465.6054,
+      "eval_samples_per_second": 33.941,
+      "eval_steps_per_second": 0.709,
+      "eval_wer": 0.18734028538323572,
+      "step": 5400
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.1016808121151608e-05,
+      "loss": 0.1251,
+      "step": 6480
+    },
+    {
+      "epoch": 0.99,
+      "eval_loss": 0.11952650547027588,
+      "eval_runtime": 466.6012,
+      "eval_samples_per_second": 33.868,
+      "eval_steps_per_second": 0.707,
+      "eval_wer": 0.1837848991317874,
+      "step": 6480
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.219504077217509e-06,
+      "loss": 0.118,
+      "step": 7560
+    },
+    {
+      "epoch": 1.15,
+      "eval_loss": 0.12100061029195786,
+      "eval_runtime": 469.3607,
+      "eval_samples_per_second": 33.669,
+      "eval_steps_per_second": 0.703,
+      "eval_wer": 0.18134057108391663,
+      "step": 7560
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.422200033283409e-06,
+      "loss": 0.1156,
+      "step": 8640
+    },
+    {
+      "epoch": 1.32,
+      "eval_loss": 0.12037315964698792,
+      "eval_runtime": 463.2023,
+      "eval_samples_per_second": 34.117,
+      "eval_steps_per_second": 0.712,
+      "eval_wer": 0.17957875021824357,
+      "step": 8640
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.62656015976036e-06,
+      "loss": 0.1152,
+      "step": 9720
+    },
+    {
+      "epoch": 1.48,
+      "eval_loss": 0.11911876499652863,
+      "eval_runtime": 465.9842,
+      "eval_samples_per_second": 33.913,
+      "eval_steps_per_second": 0.708,
+      "eval_wer": 0.17893592368617367,
+      "step": 9720
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.829256115826261e-06,
+      "loss": 0.1132,
+      "step": 10800
+    },
+    {
+      "epoch": 1.65,
+      "eval_loss": 0.11926092952489853,
+      "eval_runtime": 466.8163,
+      "eval_samples_per_second": 33.853,
+      "eval_steps_per_second": 0.707,
+      "eval_wer": 0.1785629255749726,
+      "step": 10800
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 2.033616242303212e-06,
+      "loss": 0.1133,
+      "step": 11880
+    },
+    {
+      "epoch": 1.81,
+      "eval_loss": 0.11755573004484177,
+      "eval_runtime": 568.3919,
+      "eval_samples_per_second": 27.803,
+      "eval_steps_per_second": 0.581,
+      "eval_wer": 0.17685665761947844,
+      "step": 11880
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.36312198369113e-07,
+      "loss": 0.111,
+      "step": 12960
+    },
+    {
+      "epoch": 1.98,
+      "eval_loss": 0.11704171448945999,
+      "eval_runtime": 495.4724,
+      "eval_samples_per_second": 31.895,
+      "eval_steps_per_second": 0.666,
+      "eval_wer": 0.17608685300699967,
+      "step": 12960
+    }
+  ],
+  "max_steps": 13098,
+  "num_train_epochs": 2,
+  "total_flos": 5.0362098492856254e+19,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87c4b00b8854fd0416c9bf87368a6261d74c045316d9f079f529d772cdd544de
+size 2991