patrickvonplaten commited on
Commit
b3efa5d
1 Parent(s): 96e6331
config.json ADDED
@@ -0,0 +1,231 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SpeechEncoderDecoderModel"
4
+ ],
5
+ "bos_token_id": 0,
6
+ "decoder": {
7
+ "_name_or_path": "",
8
+ "activation_dropout": 0.1,
9
+ "activation_function": "relu",
10
+ "add_cross_attention": true,
11
+ "architectures": [
12
+ "Speech2TextForConditionalGeneration"
13
+ ],
14
+ "attention_dropout": 0.1,
15
+ "bad_words_ids": null,
16
+ "bos_token_id": 0,
17
+ "chunk_size_feed_forward": 0,
18
+ "classifier_dropout": 0.0,
19
+ "conv_channels": 1024,
20
+ "conv_kernel_sizes": [
21
+ 5,
22
+ 5
23
+ ],
24
+ "d_model": 256,
25
+ "decoder_attention_heads": 4,
26
+ "decoder_ffn_dim": 2048,
27
+ "decoder_layerdrop": 0.0,
28
+ "decoder_layers": 7,
29
+ "decoder_start_token_id": 2,
30
+ "diversity_penalty": 0.0,
31
+ "do_sample": false,
32
+ "dropout": 0.1,
33
+ "early_stopping": true,
34
+ "encoder_attention_heads": 4,
35
+ "encoder_ffn_dim": 2048,
36
+ "encoder_layerdrop": 0.0,
37
+ "encoder_layers": 12,
38
+ "encoder_no_repeat_ngram_size": 0,
39
+ "eos_token_id": 2,
40
+ "finetuning_task": null,
41
+ "forced_bos_token_id": null,
42
+ "forced_eos_token_id": null,
43
+ "gradient_checkpointing": false,
44
+ "id2label": {
45
+ "0": "LABEL_0",
46
+ "1": "LABEL_1"
47
+ },
48
+ "init_std": 0.02,
49
+ "input_channels": 1,
50
+ "input_feat_per_channel": 80,
51
+ "is_decoder": true,
52
+ "is_encoder_decoder": true,
53
+ "label2id": {
54
+ "LABEL_0": 0,
55
+ "LABEL_1": 1
56
+ },
57
+ "length_penalty": 1.0,
58
+ "max_length": 200,
59
+ "max_source_positions": 6000,
60
+ "max_target_positions": 1024,
61
+ "min_length": 0,
62
+ "model_type": "speech_to_text_2",
63
+ "no_repeat_ngram_size": 0,
64
+ "num_beam_groups": 1,
65
+ "num_beams": 5,
66
+ "num_conv_layers": 2,
67
+ "num_hidden_layers": 7,
68
+ "num_return_sequences": 1,
69
+ "output_attentions": false,
70
+ "output_hidden_states": false,
71
+ "output_scores": false,
72
+ "pad_token_id": 1,
73
+ "prefix": null,
74
+ "problem_type": null,
75
+ "pruned_heads": {},
76
+ "remove_invalid_values": false,
77
+ "repetition_penalty": 1.0,
78
+ "return_dict": true,
79
+ "return_dict_in_generate": false,
80
+ "scale_embedding": true,
81
+ "sep_token_id": null,
82
+ "task_specific_params": null,
83
+ "temperature": 1.0,
84
+ "tie_encoder_decoder": false,
85
+ "tie_word_embeddings": false,
86
+ "tokenizer_class": null,
87
+ "top_k": 50,
88
+ "top_p": 1.0,
89
+ "torch_dtype": null,
90
+ "torchscript": false,
91
+ "transformers_version": "4.10.0.dev0",
92
+ "use_bfloat16": false,
93
+ "use_cache": true,
94
+ "vocab_size": 10218
95
+ },
96
+ "encoder": {
97
+ "_name_or_path": "",
98
+ "activation_dropout": 0.1,
99
+ "add_cross_attention": false,
100
+ "apply_spec_augment": true,
101
+ "architectures": [
102
+ "Wav2Vec2ForPreTraining"
103
+ ],
104
+ "attention_dropout": 0.1,
105
+ "bad_words_ids": null,
106
+ "bos_token_id": 1,
107
+ "chunk_size_feed_forward": 0,
108
+ "codevector_dim": 768,
109
+ "contrastive_logits_temperature": 0.1,
110
+ "conv_bias": true,
111
+ "conv_dim": [
112
+ 512,
113
+ 512,
114
+ 512,
115
+ 512,
116
+ 512,
117
+ 512,
118
+ 512
119
+ ],
120
+ "conv_kernel": [
121
+ 10,
122
+ 3,
123
+ 3,
124
+ 3,
125
+ 3,
126
+ 2,
127
+ 2
128
+ ],
129
+ "conv_stride": [
130
+ 5,
131
+ 2,
132
+ 2,
133
+ 2,
134
+ 2,
135
+ 2,
136
+ 2
137
+ ],
138
+ "ctc_loss_reduction": "sum",
139
+ "ctc_zero_infinity": false,
140
+ "decoder_start_token_id": null,
141
+ "diversity_loss_weight": 0.1,
142
+ "diversity_penalty": 0.0,
143
+ "do_sample": false,
144
+ "do_stable_layer_norm": true,
145
+ "early_stopping": false,
146
+ "encoder_no_repeat_ngram_size": 0,
147
+ "eos_token_id": 2,
148
+ "feat_extract_activation": "gelu",
149
+ "feat_extract_dropout": 0.0,
150
+ "feat_extract_norm": "layer",
151
+ "feat_proj_dropout": 0.1,
152
+ "feat_quantizer_dropout": 0.0,
153
+ "final_dropout": 0.1,
154
+ "finetuning_task": null,
155
+ "forced_bos_token_id": null,
156
+ "forced_eos_token_id": null,
157
+ "gradient_checkpointing": false,
158
+ "hidden_act": "gelu",
159
+ "hidden_dropout": 0.1,
160
+ "hidden_dropout_prob": 0.1,
161
+ "hidden_size": 1024,
162
+ "id2label": {
163
+ "0": "LABEL_0",
164
+ "1": "LABEL_1"
165
+ },
166
+ "initializer_range": 0.02,
167
+ "intermediate_size": 4096,
168
+ "is_decoder": false,
169
+ "is_encoder_decoder": false,
170
+ "label2id": {
171
+ "LABEL_0": 0,
172
+ "LABEL_1": 1
173
+ },
174
+ "layer_norm_eps": 1e-05,
175
+ "layerdrop": 0.1,
176
+ "length_penalty": 1.0,
177
+ "mask_feature_length": 10,
178
+ "mask_feature_prob": 0.0,
179
+ "mask_time_length": 10,
180
+ "mask_time_prob": 0.05,
181
+ "max_length": 20,
182
+ "min_length": 0,
183
+ "model_type": "wav2vec2",
184
+ "no_repeat_ngram_size": 0,
185
+ "num_attention_heads": 16,
186
+ "num_beam_groups": 1,
187
+ "num_beams": 1,
188
+ "num_codevector_groups": 2,
189
+ "num_codevectors_per_group": 320,
190
+ "num_conv_pos_embedding_groups": 16,
191
+ "num_conv_pos_embeddings": 128,
192
+ "num_feat_extract_layers": 7,
193
+ "num_hidden_layers": 24,
194
+ "num_negatives": 100,
195
+ "num_return_sequences": 1,
196
+ "output_attentions": false,
197
+ "output_hidden_states": false,
198
+ "output_scores": false,
199
+ "pad_token_id": 0,
200
+ "prefix": null,
201
+ "problem_type": null,
202
+ "proj_codevector_dim": 768,
203
+ "pruned_heads": {},
204
+ "remove_invalid_values": false,
205
+ "repetition_penalty": 1.0,
206
+ "return_dict": true,
207
+ "return_dict_in_generate": false,
208
+ "sep_token_id": null,
209
+ "task_specific_params": null,
210
+ "temperature": 1.0,
211
+ "tie_encoder_decoder": false,
212
+ "tie_word_embeddings": true,
213
+ "tokenizer_class": null,
214
+ "top_k": 50,
215
+ "top_p": 1.0,
216
+ "torch_dtype": null,
217
+ "torchscript": false,
218
+ "transformers_version": "4.10.0.dev0",
219
+ "use_bfloat16": false,
220
+ "vocab_size": 32
221
+ },
222
+ "eos_token_id": 2,
223
+ "feature_extractor_type": "wav2vec2",
224
+ "is_encoder_decoder": true,
225
+ "model_type": "speech-encoder-decoder",
226
+ "pad_token_id": 1,
227
+ "tie_word_embeddings": false,
228
+ "tokenizer_class": "speech_to_text_2",
229
+ "torch_dtype": "float32",
230
+ "transformers_version": null
231
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c95f026fb4b64afdb66c34ca09a44b1db54dcda9eb8c438ec12ca4e1f4f144b
3
+ size 1329232088
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "tokenizer_class": "Speech2Text2Tokenizer"}
vocab.json ADDED
The diff for this file is too large to render. See raw diff