jonatasgrosman commited on
Commit
9081f4a
1 Parent(s): 5184a49

jonatasgrosman/exp_w2v2r_es_xls-r_age_teens-2_sixties-8_s772

Browse files
.gitattributes CHANGED
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ cv7_transcriptions.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - es
4
+ license: apache-2.0
5
+ tags:
6
+ - automatic-speech-recognition
7
+ - es
8
+ datasets:
9
+ - mozilla-foundation/common_voice_7_0
10
+ ---
11
+ # exp_w2v2r_es_xls-r_age_teens-2_sixties-8_s772
12
+
13
+ Fine-tuned [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) for speech recognition using the train split of [Common Voice 7.0 (es)](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0).
14
+ When using this model, make sure that your speech input is sampled at 16kHz.
15
+
16
+ This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool.
17
+
config.json ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-xls-r-300m",
3
+ "activation_dropout": 0.05,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Wav2Vec2ForCTC"
10
+ ],
11
+ "attention_dropout": 0.05,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": true,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "mean",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": true,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_dropout": 0.0,
51
+ "feat_extract_norm": "layer",
52
+ "feat_proj_dropout": 0.05,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.05,
55
+ "gradient_checkpointing": false,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.05,
58
+ "hidden_size": 1024,
59
+ "initializer_range": 0.02,
60
+ "intermediate_size": 4096,
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.05,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "wav2vec2",
70
+ "num_adapter_layers": 3,
71
+ "num_attention_heads": 16,
72
+ "num_codevector_groups": 2,
73
+ "num_codevectors_per_group": 320,
74
+ "num_conv_pos_embedding_groups": 16,
75
+ "num_conv_pos_embeddings": 128,
76
+ "num_feat_extract_layers": 7,
77
+ "num_hidden_layers": 24,
78
+ "num_negatives": 100,
79
+ "output_hidden_size": 1024,
80
+ "pad_token_id": 0,
81
+ "proj_codevector_dim": 768,
82
+ "tdnn_dilation": [
83
+ 1,
84
+ 2,
85
+ 3,
86
+ 1,
87
+ 1
88
+ ],
89
+ "tdnn_dim": [
90
+ 512,
91
+ 512,
92
+ 512,
93
+ 512,
94
+ 1500
95
+ ],
96
+ "tdnn_kernel": [
97
+ 5,
98
+ 3,
99
+ 3,
100
+ 1,
101
+ 1
102
+ ],
103
+ "torch_dtype": "float32",
104
+ "transformers_version": "4.18.0",
105
+ "use_weighted_layer_sum": false,
106
+ "vocab_size": 41,
107
+ "xvector_output_dim": 512
108
+ }
cv7_evaluation.json ADDED
@@ -0,0 +1,300 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "validation": [
3
+ {
4
+ "subset_type": "all",
5
+ "subset": "all",
6
+ "cer": 0.039691048221378714,
7
+ "wer": 0.14549307490196597
8
+ },
9
+ {
10
+ "subset_type": "age",
11
+ "subset": "",
12
+ "cer": 0.039599512499291424,
13
+ "wer": 0.14495660821596443
14
+ },
15
+ {
16
+ "subset_type": "age",
17
+ "subset": "twenties",
18
+ "cer": 0.04166974328681484,
19
+ "wer": 0.15342203925604647
20
+ },
21
+ {
22
+ "subset_type": "age",
23
+ "subset": "fourties",
24
+ "cer": 0.033323998879865584,
25
+ "wer": 0.12843651626442812
26
+ },
27
+ {
28
+ "subset_type": "age",
29
+ "subset": "thirties",
30
+ "cer": 0.03983683965505775,
31
+ "wer": 0.14880401744416547
32
+ },
33
+ {
34
+ "subset_type": "age",
35
+ "subset": "fifties",
36
+ "cer": 0.03665728454630746,
37
+ "wer": 0.1312199036918138
38
+ },
39
+ {
40
+ "subset_type": "age",
41
+ "subset": "sixties",
42
+ "cer": 0.0482547216454069,
43
+ "wer": 0.15186915887850466
44
+ },
45
+ {
46
+ "subset_type": "age",
47
+ "subset": "seventies",
48
+ "cer": 0.03387096774193549,
49
+ "wer": 0.12264150943396226
50
+ },
51
+ {
52
+ "subset_type": "age",
53
+ "subset": "teens",
54
+ "cer": 0.0418615866891729,
55
+ "wer": 0.1572742955721679
56
+ },
57
+ {
58
+ "subset_type": "gender",
59
+ "subset": "",
60
+ "cer": 0.03953642946034619,
61
+ "wer": 0.14475815685243454
62
+ },
63
+ {
64
+ "subset_type": "gender",
65
+ "subset": "other",
66
+ "cer": 0.04891304347826087,
67
+ "wer": 0.17540687160940324
68
+ },
69
+ {
70
+ "subset_type": "gender",
71
+ "subset": "male",
72
+ "cer": 0.04123544316922539,
73
+ "wer": 0.15041264675113333
74
+ },
75
+ {
76
+ "subset_type": "gender",
77
+ "subset": "female",
78
+ "cer": 0.03596769676033984,
79
+ "wer": 0.13760577056457207
80
+ },
81
+ {
82
+ "subset_type": "accent",
83
+ "subset": "",
84
+ "cer": 0.03954381195504426,
85
+ "wer": 0.14490840859941484
86
+ },
87
+ {
88
+ "subset_type": "accent",
89
+ "subset": "caribe",
90
+ "cer": 0.03907637655417407,
91
+ "wer": 0.14233210825258938
92
+ },
93
+ {
94
+ "subset_type": "accent",
95
+ "subset": "mexicano",
96
+ "cer": 0.03926864741325577,
97
+ "wer": 0.14761575673807878
98
+ },
99
+ {
100
+ "subset_type": "accent",
101
+ "subset": "surpeninsular",
102
+ "cer": 0.04100227790432802,
103
+ "wer": 0.14733542319749215
104
+ },
105
+ {
106
+ "subset_type": "accent",
107
+ "subset": "chileno",
108
+ "cer": 0.05033051142293865,
109
+ "wer": 0.1782042494859493
110
+ },
111
+ {
112
+ "subset_type": "accent",
113
+ "subset": "canario",
114
+ "cer": 0.03961848862802641,
115
+ "wer": 0.1651376146788991
116
+ },
117
+ {
118
+ "subset_type": "accent",
119
+ "subset": "americacentral",
120
+ "cer": 0.0435072989218586,
121
+ "wer": 0.15011547344110854
122
+ },
123
+ {
124
+ "subset_type": "accent",
125
+ "subset": "nortepeninsular",
126
+ "cer": 0.03556445461479787,
127
+ "wer": 0.13333333333333333
128
+ },
129
+ {
130
+ "subset_type": "accent",
131
+ "subset": "andino",
132
+ "cer": 0.037315713085325795,
133
+ "wer": 0.14214236979758926
134
+ },
135
+ {
136
+ "subset_type": "accent",
137
+ "subset": "rioplatense",
138
+ "cer": 0.03977677511279981,
139
+ "wer": 0.14577363896848136
140
+ },
141
+ {
142
+ "subset_type": "accent",
143
+ "subset": "centrosurpeninsular",
144
+ "cer": 0.04812234766594603,
145
+ "wer": 0.16494845360824742
146
+ }
147
+ ],
148
+ "test": [
149
+ {
150
+ "subset_type": "all",
151
+ "subset": "all",
152
+ "cer": 0.045896383065992176,
153
+ "wer": 0.16060491942996452
154
+ },
155
+ {
156
+ "subset_type": "age",
157
+ "subset": "",
158
+ "cer": 0.046012698873052704,
159
+ "wer": 0.16073091247672253
160
+ },
161
+ {
162
+ "subset_type": "age",
163
+ "subset": "twenties",
164
+ "cer": 0.04660719655206976,
165
+ "wer": 0.16798656107511398
166
+ },
167
+ {
168
+ "subset_type": "age",
169
+ "subset": "fourties",
170
+ "cer": 0.037473828141691236,
171
+ "wer": 0.13667953667953667
172
+ },
173
+ {
174
+ "subset_type": "age",
175
+ "subset": "thirties",
176
+ "cer": 0.04242727028169555,
177
+ "wer": 0.1560919540229885
178
+ },
179
+ {
180
+ "subset_type": "age",
181
+ "subset": "fifties",
182
+ "cer": 0.04502111428960632,
183
+ "wer": 0.14129995962858297
184
+ },
185
+ {
186
+ "subset_type": "age",
187
+ "subset": "sixties",
188
+ "cer": 0.04586086280606296,
189
+ "wer": 0.16032295271049596
190
+ },
191
+ {
192
+ "subset_type": "age",
193
+ "subset": "seventies",
194
+ "cer": 0.1101871101871102,
195
+ "wer": 0.2926829268292683
196
+ },
197
+ {
198
+ "subset_type": "age",
199
+ "subset": "teens",
200
+ "cer": 0.055941228011053445,
201
+ "wer": 0.1908831908831909
202
+ },
203
+ {
204
+ "subset_type": "age",
205
+ "subset": "eighties",
206
+ "cer": 0.013422818791946308,
207
+ "wer": 0.07407407407407407
208
+ },
209
+ {
210
+ "subset_type": "gender",
211
+ "subset": "",
212
+ "cer": 0.046063490299123815,
213
+ "wer": 0.16090778823438287
214
+ },
215
+ {
216
+ "subset_type": "gender",
217
+ "subset": "other",
218
+ "cer": 0.07388809182209469,
219
+ "wer": 0.22608695652173913
220
+ },
221
+ {
222
+ "subset_type": "gender",
223
+ "subset": "male",
224
+ "cer": 0.04459669704202623,
225
+ "wer": 0.15797386376778072
226
+ },
227
+ {
228
+ "subset_type": "gender",
229
+ "subset": "female",
230
+ "cer": 0.044705136334812934,
231
+ "wer": 0.1587737843551797
232
+ },
233
+ {
234
+ "subset_type": "accent",
235
+ "subset": "",
236
+ "cer": 0.04629755376071943,
237
+ "wer": 0.16184838506789004
238
+ },
239
+ {
240
+ "subset_type": "accent",
241
+ "subset": "caribe",
242
+ "cer": 0.04942327347057965,
243
+ "wer": 0.16680923866552608
244
+ },
245
+ {
246
+ "subset_type": "accent",
247
+ "subset": "surpeninsular",
248
+ "cer": 0.03672869735553379,
249
+ "wer": 0.13777777777777778
250
+ },
251
+ {
252
+ "subset_type": "accent",
253
+ "subset": "mexicano",
254
+ "cer": 0.04210045662100457,
255
+ "wer": 0.15137101287073307
256
+ },
257
+ {
258
+ "subset_type": "accent",
259
+ "subset": "chileno",
260
+ "cer": 0.038402220851326344,
261
+ "wer": 0.13815187557182068
262
+ },
263
+ {
264
+ "subset_type": "accent",
265
+ "subset": "canario",
266
+ "cer": 0.05555555555555555,
267
+ "wer": 0.21019108280254778
268
+ },
269
+ {
270
+ "subset_type": "accent",
271
+ "subset": "americacentral",
272
+ "cer": 0.041551972735614104,
273
+ "wer": 0.14576802507836992
274
+ },
275
+ {
276
+ "subset_type": "accent",
277
+ "subset": "nortepeninsular",
278
+ "cer": 0.040202041026698276,
279
+ "wer": 0.14146933819064966
280
+ },
281
+ {
282
+ "subset_type": "accent",
283
+ "subset": "andino",
284
+ "cer": 0.038625894888728564,
285
+ "wer": 0.14266622118275976
286
+ },
287
+ {
288
+ "subset_type": "accent",
289
+ "subset": "rioplatense",
290
+ "cer": 0.05820707070707071,
291
+ "wer": 0.1871737509321402
292
+ },
293
+ {
294
+ "subset_type": "accent",
295
+ "subset": "centrosurpeninsular",
296
+ "cer": 0.03195860599604322,
297
+ "wer": 0.11696428571428572
298
+ }
299
+ ]
300
+ }
cv7_transcriptions.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f3bfde59903e571da37822885d866c95b058e2ca54a777d5ecc775f50b39c7e
3
+ size 125421244
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41e68a21ced703adfad8834d51b3827cba3cbb2864040c2e8a9175e848ad6bcc
3
+ size 1262066801
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>", "do_lower_case": false, "word_delimiter_token": "|", "replace_word_delimiter_char": " ", "special_tokens_map_file": null, "name_or_path": "/tmp/tmpj8d9jtml", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2Processor"}
vocab.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "'": 5, "-": 6, "a": 7, "b": 8, "c": 9, "d": 10, "e": 11, "f": 12, "g": 13, "h": 14, "i": 15, "j": 16, "k": 17, "l": 18, "m": 19, "n": 20, "o": 21, "p": 22, "q": 23, "r": 24, "s": 25, "t": 26, "u": 27, "v": 28, "w": 29, "x": 30, "y": 31, "z": 32, "á": 33, "é": 34, "í": 35, "ñ": 36, "ó": 37, "ö": 38, "ú": 39, "ü": 40}