innovation64 commited on
Commit
35cff7d
1 Parent(s): f06fe23

End of training

Browse files
Files changed (2) hide show
  1. README.md +78 -0
  2. generation_config.json +253 -0
README.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: openai/whisper-tiny
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - PolyAI/minds14
8
+ metrics:
9
+ - wer
10
+ model-index:
11
+ - name: whisper-tiny-dv
12
+ results:
13
+ - task:
14
+ name: Automatic Speech Recognition
15
+ type: automatic-speech-recognition
16
+ dataset:
17
+ name: PolyAI/minds14
18
+ type: PolyAI/minds14
19
+ config: en-US
20
+ split: train
21
+ args: en-US
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 34.179456906729634
26
+ ---
27
+
28
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
+ should probably proofread and complete it, then remove this comment. -->
30
+
31
+ # whisper-tiny-dv
32
+
33
+ This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 0.6498
36
+ - Wer Ortho: 34.6083
37
+ - Wer: 34.1795
38
+
39
+ ## Model description
40
+
41
+ More information needed
42
+
43
+ ## Intended uses & limitations
44
+
45
+ More information needed
46
+
47
+ ## Training and evaluation data
48
+
49
+ More information needed
50
+
51
+ ## Training procedure
52
+
53
+ ### Training hyperparameters
54
+
55
+ The following hyperparameters were used during training:
56
+ - learning_rate: 1e-05
57
+ - train_batch_size: 16
58
+ - eval_batch_size: 16
59
+ - seed: 42
60
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
+ - lr_scheduler_type: constant_with_warmup
62
+ - lr_scheduler_warmup_steps: 50
63
+ - training_steps: 500
64
+ - mixed_precision_training: Native AMP
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
69
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
70
+ | 0.0007 | 17.86 | 500 | 0.6498 | 34.6083 | 34.1795 |
71
+
72
+
73
+ ### Framework versions
74
+
75
+ - Transformers 4.35.0.dev0
76
+ - Pytorch 2.1.0+cu118
77
+ - Datasets 2.14.6
78
+ - Tokenizers 0.14.1
generation_config.json ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 2,
5
+ 2
6
+ ],
7
+ [
8
+ 3,
9
+ 0
10
+ ],
11
+ [
12
+ 3,
13
+ 2
14
+ ],
15
+ [
16
+ 3,
17
+ 3
18
+ ],
19
+ [
20
+ 3,
21
+ 4
22
+ ],
23
+ [
24
+ 3,
25
+ 5
26
+ ]
27
+ ],
28
+ "begin_suppress_tokens": [
29
+ 220,
30
+ 50257
31
+ ],
32
+ "bos_token_id": 50257,
33
+ "decoder_start_token_id": 50258,
34
+ "eos_token_id": 50257,
35
+ "forced_decoder_ids": [
36
+ [
37
+ 1,
38
+ 50259
39
+ ],
40
+ [
41
+ 2,
42
+ 50359
43
+ ],
44
+ [
45
+ 3,
46
+ 50363
47
+ ]
48
+ ],
49
+ "is_multilingual": true,
50
+ "lang_to_id": {
51
+ "<|af|>": 50327,
52
+ "<|am|>": 50334,
53
+ "<|ar|>": 50272,
54
+ "<|as|>": 50350,
55
+ "<|az|>": 50304,
56
+ "<|ba|>": 50355,
57
+ "<|be|>": 50330,
58
+ "<|bg|>": 50292,
59
+ "<|bn|>": 50302,
60
+ "<|bo|>": 50347,
61
+ "<|br|>": 50309,
62
+ "<|bs|>": 50315,
63
+ "<|ca|>": 50270,
64
+ "<|cs|>": 50283,
65
+ "<|cy|>": 50297,
66
+ "<|da|>": 50285,
67
+ "<|de|>": 50261,
68
+ "<|el|>": 50281,
69
+ "<|en|>": 50259,
70
+ "<|es|>": 50262,
71
+ "<|et|>": 50307,
72
+ "<|eu|>": 50310,
73
+ "<|fa|>": 50300,
74
+ "<|fi|>": 50277,
75
+ "<|fo|>": 50338,
76
+ "<|fr|>": 50265,
77
+ "<|gl|>": 50319,
78
+ "<|gu|>": 50333,
79
+ "<|haw|>": 50352,
80
+ "<|ha|>": 50354,
81
+ "<|he|>": 50279,
82
+ "<|hi|>": 50276,
83
+ "<|hr|>": 50291,
84
+ "<|ht|>": 50339,
85
+ "<|hu|>": 50286,
86
+ "<|hy|>": 50312,
87
+ "<|id|>": 50275,
88
+ "<|is|>": 50311,
89
+ "<|it|>": 50274,
90
+ "<|ja|>": 50266,
91
+ "<|jw|>": 50356,
92
+ "<|ka|>": 50329,
93
+ "<|kk|>": 50316,
94
+ "<|km|>": 50323,
95
+ "<|kn|>": 50306,
96
+ "<|ko|>": 50264,
97
+ "<|la|>": 50294,
98
+ "<|lb|>": 50345,
99
+ "<|ln|>": 50353,
100
+ "<|lo|>": 50336,
101
+ "<|lt|>": 50293,
102
+ "<|lv|>": 50301,
103
+ "<|mg|>": 50349,
104
+ "<|mi|>": 50295,
105
+ "<|mk|>": 50308,
106
+ "<|ml|>": 50296,
107
+ "<|mn|>": 50314,
108
+ "<|mr|>": 50320,
109
+ "<|ms|>": 50282,
110
+ "<|mt|>": 50343,
111
+ "<|my|>": 50346,
112
+ "<|ne|>": 50313,
113
+ "<|nl|>": 50271,
114
+ "<|nn|>": 50342,
115
+ "<|no|>": 50288,
116
+ "<|oc|>": 50328,
117
+ "<|pa|>": 50321,
118
+ "<|pl|>": 50269,
119
+ "<|ps|>": 50340,
120
+ "<|pt|>": 50267,
121
+ "<|ro|>": 50284,
122
+ "<|ru|>": 50263,
123
+ "<|sa|>": 50344,
124
+ "<|sd|>": 50332,
125
+ "<|si|>": 50322,
126
+ "<|sk|>": 50298,
127
+ "<|sl|>": 50305,
128
+ "<|sn|>": 50324,
129
+ "<|so|>": 50326,
130
+ "<|sq|>": 50317,
131
+ "<|sr|>": 50303,
132
+ "<|su|>": 50357,
133
+ "<|sv|>": 50273,
134
+ "<|sw|>": 50318,
135
+ "<|ta|>": 50287,
136
+ "<|te|>": 50299,
137
+ "<|tg|>": 50331,
138
+ "<|th|>": 50289,
139
+ "<|tk|>": 50341,
140
+ "<|tl|>": 50348,
141
+ "<|tr|>": 50268,
142
+ "<|tt|>": 50351,
143
+ "<|uk|>": 50280,
144
+ "<|ur|>": 50290,
145
+ "<|uz|>": 50337,
146
+ "<|vi|>": 50278,
147
+ "<|yi|>": 50335,
148
+ "<|yo|>": 50325,
149
+ "<|zh|>": 50260
150
+ },
151
+ "language": "english",
152
+ "max_initial_timestamp_index": 1,
153
+ "max_length": 448,
154
+ "no_timestamps_token_id": 50363,
155
+ "pad_token_id": 50257,
156
+ "return_timestamps": false,
157
+ "suppress_tokens": [
158
+ 1,
159
+ 2,
160
+ 7,
161
+ 8,
162
+ 9,
163
+ 10,
164
+ 14,
165
+ 25,
166
+ 26,
167
+ 27,
168
+ 28,
169
+ 29,
170
+ 31,
171
+ 58,
172
+ 59,
173
+ 60,
174
+ 61,
175
+ 62,
176
+ 63,
177
+ 90,
178
+ 91,
179
+ 92,
180
+ 93,
181
+ 359,
182
+ 503,
183
+ 522,
184
+ 542,
185
+ 873,
186
+ 893,
187
+ 902,
188
+ 918,
189
+ 922,
190
+ 931,
191
+ 1350,
192
+ 1853,
193
+ 1982,
194
+ 2460,
195
+ 2627,
196
+ 3246,
197
+ 3253,
198
+ 3268,
199
+ 3536,
200
+ 3846,
201
+ 3961,
202
+ 4183,
203
+ 4667,
204
+ 6585,
205
+ 6647,
206
+ 7273,
207
+ 9061,
208
+ 9383,
209
+ 10428,
210
+ 10929,
211
+ 11938,
212
+ 12033,
213
+ 12331,
214
+ 12562,
215
+ 13793,
216
+ 14157,
217
+ 14635,
218
+ 15265,
219
+ 15618,
220
+ 16553,
221
+ 16604,
222
+ 18362,
223
+ 18956,
224
+ 20075,
225
+ 21675,
226
+ 22520,
227
+ 26130,
228
+ 26161,
229
+ 26435,
230
+ 28279,
231
+ 29464,
232
+ 31650,
233
+ 32302,
234
+ 32470,
235
+ 36865,
236
+ 42863,
237
+ 47425,
238
+ 49870,
239
+ 50254,
240
+ 50258,
241
+ 50358,
242
+ 50359,
243
+ 50360,
244
+ 50361,
245
+ 50362
246
+ ],
247
+ "task": "transcribe",
248
+ "task_to_id": {
249
+ "transcribe": 50359,
250
+ "translate": 50358
251
+ },
252
+ "transformers_version": "4.35.0.dev0"
253
+ }