first

config.json

@@ -0,0 +1,105 @@
+{
+  "activation_dropout": 0.1,
+  "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": "sum",
+  "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": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "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,
+  "model_type": "wav2vec2",
+  "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": 0,
+  "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": "4.18.0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 98,
+  "xvector_output_dim": 512
+}

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2Processor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

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": "|", "replace_word_delimiter_char": " ", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2Processor"}

vocab.json

@@ -0,0 +1 @@
+{"<s>": 1, "<pad>": 0, "</s>": 2, "<unk>": 3, "|": 4, "N": 5, "H": 6, "I": 7, "T": 8, "C": 9, "G": 10, "M": 11, "A": 12, "Đ": 13, "À": 14, "U": 15, "L": 16, "Á": 17, "Y": 18, "B": 19, "O": 20, "Ư": 21, "V": 22, "K": 23, "R": 24, "Ô": 25, "Ạ": 26, "S": 27, "P": 28, "Ó": 29, "Ê": 30, "D": 31, "Ì": 32, "Ệ": 33, "Ờ": 34, "Ả": 35, "Ế": 36, "Ố": 37, "Ị": 38, "Ộ": 39, "E": 40, "Â": 41, "Ơ": 42, "Ấ": 43, "Ă": 44, "Ề": 45, "Ồ": 46, "Ớ": 47, "Ể": 48, "Ú": 49, "Ợ": 50, "X": 51, "Ậ": 52, "Í": 53, "Ủ": 54, "Ầ": 55, "Q": 56, "Ở": 57, "Ắ": 58, "Ã": 59, "Ọ": 60, "Ò": 61, "Ự": 62, "Ữ": 63, "Ứ": 64, "Ụ": 65, "Ừ": 66, "Ù": 67, "Ặ": 68, "Ũ": 69, "Ổ": 70, "Ỏ": 71, "Ỉ": 72, "Ý": 73, "Ẩ": 74, "Ử": 75, "Ẽ": 76, "Ỗ": 77, "É": 78, "Ễ": 79, "Ằ": 80, "Ẫ": 81, "È": 82, "Ĩ": 83, "Ẹ": 84, "Ẻ": 85, "Ẳ": 86, "Ỡ": 87, "Ỹ": 88, "Õ": 89, "Ỳ": 90, "Ỷ": 91, "Ẵ": 92, "Ỵ": 93, "F": 94, "W": 95, "Z": 96, "J": 97}