update

added_tokens.json

@@ -0,0 +1 @@
+{"<s>": 107, "</s>": 108}

config.json

@@ -0,0 +1,76 @@
+{
+  "_name_or_path": "/content/drive/MyDrive/Speech-To-Text-2021/XLSR-Wav2Vec2/2021-05-29 12:40:04.541683/checkpoint-75",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "conv_bias": true,
+  "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": true,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.0,
+  "final_dropout": 0.0,
+  "gradient_checkpointing": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "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_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "num_attention_heads": 16,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "pad_token_id": 106,
+  "transformers_version": "4.4.0",
+  "vocab_size": 107
+}

preprocessor_config.json

@@ -0,0 +1,8 @@
+{
+  "do_normalize": true,
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fade7a3b8bb9018ff522b8e7a49e1ee2b795bd4243903aeb0ca28991c0e9c3c
+size 623

special_tokens_map.json

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

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": "|", "special_tokens_map_file": "/root/.cache/huggingface/transformers/8892bd2ba1de4982047602e38927ac6f215a3c401e48d5ed39a5341ed475633e.a21d51735cf8667bcd610f057e88548d5d6a381401f6b4501a8bc6c1a9dc8498", "tokenizer_file": null, "name_or_path": "/content/drive/MyDrive/Speech-To-Text-2021/XLSR-Wav2Vec2/2021-05-29 12:40:04.541683/checkpoint-75"}

trainer_state.json

@@ -0,0 +1,32 @@
+{
+  "best_metric": 0.06104659661650658,
+  "best_model_checkpoint": "/content/drive/MyDrive/Speech-To-Text-2021/XLSR-Wav2Vec2/2021-05-29 13:44:02.238365/checkpoint-25",
+  "epoch": 2.0,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.06104659661650658,
+      "eval_runtime": 82.3366,
+      "eval_samples_per_second": 9.23,
+      "eval_wer": 0.16317016317016317,
+      "step": 25
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.06628068536520004,
+      "eval_runtime": 83.5033,
+      "eval_samples_per_second": 9.101,
+      "eval_wer": 0.16161616161616163,
+      "step": 50
+    }
+  ],
+  "max_steps": 250,
+  "num_train_epochs": 10,
+  "total_flos": 2.075614232631e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f628a2baa271d692db28ee812eab102c39c44f9cf320b0ccd19f16f8ca01a4a3
+size 2351

vocab.json

@@ -0,0 +1 @@
+{"́": 0, "ò": 1, "ỳ": 2, "ủ": 3, "ằ": 4, "õ": 5, "ể": 6, "ỷ": 7, "ô": 8, "ẵ": 9, "ỹ": 10, "ễ": 11, "à": 12, "̣": 13, "ẫ": 14, "ẳ": 15, "h": 16, "ụ": 17, "đ": 18, "c": 19, "ồ": 20, "ộ": 21, "í": 22, "j": 23, "ă": 24, "ả": 25, "ì": 26, "ỗ": 27, "ǀ": 28, "o": 29, "̃": 30, "ặ": 31, "ẩ": 32, "ỉ": 33, "v": 34, "l": 35, "ĩ": 36, "ẻ": 37, "ừ": 38, "ầ": 39, "q": 40, "n": 41, "â": 42, "ử": 43, "m": 44, "ẽ": 45, "ọ": 46, "b": 47, "z": 48, "ế": 49, "“": 50, "ỵ": 51, "ệ": 52, "p": 53, "”": 54, "è": 55, "ã": 56, "ứ": 57, "d": 58, "x": 59, "ậ": 60, "w": 61, "ắ": 62, "t": 63, "ú": 64, "ỡ": 65, "ý": 66, "̉": 67, "ố": 68, "ợ": 69, "i": 70, "ạ": 71, "\\": 72, "é": 73, "ê": 74, "ẹ": 75, "ổ": 76, "ự": 77, "a": 78, "–": 79, "y": 80, "ờ": 81, "ở": 82, "s": 84, "ń": 85, "ớ": 86, "e": 87, "̀": 88, "ấ": 89, "ơ": 90, "ề": 91, "k": 92, "g": 93, "ó": 94, "ư": 95, "ù": 96, "ị": 97, "ũ": 98, "á": 99, "r": 100, "f": 101, "u": 102, "ữ": 103, "ỏ": 104, "|": 83, "[UNK]": 105, "[PAD]": 106}