ymoslem commited on
Commit
e74ce0c
1 Parent(s): 9136b61

End of training

Browse files
README.md ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ga
4
+ - en
5
+ license: apache-2.0
6
+ base_model: openai/whisper-large
7
+ tags:
8
+ - generated_from_trainer
9
+ datasets:
10
+ - ymoslem/IWSLT2023-GA-EN
11
+ - ymoslem/FLEURS-GA-EN
12
+ - ymoslem/BitesizeIrish-GA-EN
13
+ - ymoslem/SpokenWords-GA-EN-MTed
14
+ metrics:
15
+ - bleu
16
+ - wer
17
+ model-index:
18
+ - name: Whisper Large GA-EN Speech Translation
19
+ results:
20
+ - task:
21
+ name: Automatic Speech Recognition
22
+ type: automatic-speech-recognition
23
+ dataset:
24
+ name: IWSLT-2023, FLEURS, BiteSize, SpokenWords, Tatoeba, and Wikimedia
25
+ type: ymoslem/IWSLT2023-GA-EN
26
+ metrics:
27
+ - name: Bleu
28
+ type: bleu
29
+ value: 30.16
30
+ - name: Wer
31
+ type: wer
32
+ value: 69.968482665466
33
+ ---
34
+
35
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
36
+ should probably proofread and complete it, then remove this comment. -->
37
+
38
+ # Whisper Large GA-EN Speech Translation
39
+
40
+ This model is a fine-tuned version of [openai/whisper-large](https://huggingface.co/openai/whisper-large) on the IWSLT-2023, FLEURS, BiteSize, SpokenWords, Tatoeba, and Wikimedia dataset.
41
+ It achieves the following results on the evaluation set:
42
+ - Loss: 1.1742
43
+ - Bleu: 30.16
44
+ - Chrf: 50.72
45
+ - Wer: 69.9685
46
+
47
+ ## Model description
48
+
49
+ More information needed
50
+
51
+ ## Intended uses & limitations
52
+
53
+ More information needed
54
+
55
+ ## Training and evaluation data
56
+
57
+ More information needed
58
+
59
+ ## Training procedure
60
+
61
+ ### Training hyperparameters
62
+
63
+ The following hyperparameters were used during training:
64
+ - learning_rate: 0.0001
65
+ - train_batch_size: 4
66
+ - eval_batch_size: 4
67
+ - seed: 42
68
+ - gradient_accumulation_steps: 4
69
+ - total_train_batch_size: 16
70
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
71
+ - lr_scheduler_type: linear
72
+ - lr_scheduler_warmup_steps: 0.03
73
+ - training_steps: 3000
74
+ - mixed_precision_training: Native AMP
75
+
76
+ ### Training results
77
+
78
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Chrf | Wer |
79
+ |:-------------:|:-----:|:----:|:---------------:|:-----:|:-----:|:--------:|
80
+ | 3.1833 | 0.03 | 100 | 2.5169 | 2.03 | 16.8 | 215.5786 |
81
+ | 2.7632 | 0.07 | 200 | 2.1827 | 7.81 | 24.07 | 113.1022 |
82
+ | 2.5687 | 0.1 | 300 | 2.0746 | 6.16 | 24.2 | 158.8474 |
83
+ | 2.5615 | 0.13 | 400 | 1.9379 | 8.68 | 26.18 | 120.8465 |
84
+ | 2.4554 | 0.16 | 500 | 1.8932 | 12.14 | 28.94 | 103.1067 |
85
+ | 2.3546 | 0.2 | 600 | 1.8734 | 14.34 | 29.83 | 91.5353 |
86
+ | 2.2804 | 0.23 | 700 | 1.8075 | 13.18 | 33.07 | 105.5380 |
87
+ | 2.1408 | 0.26 | 800 | 1.7034 | 13.01 | 33.0 | 89.4642 |
88
+ | 2.0411 | 0.3 | 900 | 1.6556 | 16.73 | 34.97 | 91.4453 |
89
+ | 1.7766 | 0.33 | 1000 | 1.6505 | 17.21 | 35.54 | 83.5209 |
90
+ | 1.7704 | 0.36 | 1100 | 1.5800 | 17.54 | 38.11 | 77.1724 |
91
+ | 1.6537 | 0.39 | 1200 | 1.5684 | 14.2 | 35.39 | 95.6326 |
92
+ | 1.4841 | 0.43 | 1300 | 1.4970 | 22.96 | 39.35 | 71.3643 |
93
+ | 1.641 | 0.46 | 1400 | 1.4693 | 16.3 | 37.69 | 103.7821 |
94
+ | 1.393 | 0.49 | 1500 | 1.3923 | 27.21 | 43.87 | 69.3381 |
95
+ | 1.249 | 0.53 | 1600 | 1.3876 | 23.33 | 42.26 | 76.5421 |
96
+ | 1.3385 | 0.56 | 1700 | 1.3404 | 23.86 | 42.82 | 75.0563 |
97
+ | 1.2537 | 0.59 | 1800 | 1.3226 | 17.03 | 41.72 | 100.1801 |
98
+ | 1.2891 | 0.62 | 1900 | 1.2995 | 27.26 | 43.62 | 69.1580 |
99
+ | 1.226 | 0.66 | 2000 | 1.2605 | 30.89 | 47.34 | 63.5750 |
100
+ | 1.1268 | 0.69 | 2100 | 1.2783 | 27.43 | 45.97 | 67.4921 |
101
+ | 1.0007 | 0.72 | 2200 | 1.2521 | 27.21 | 47.25 | 71.0041 |
102
+ | 0.9565 | 0.76 | 2300 | 1.2219 | 31.65 | 48.07 | 64.2053 |
103
+ | 0.9309 | 0.79 | 2400 | 1.2193 | 31.4 | 48.18 | 64.1603 |
104
+ | 0.7923 | 0.82 | 2500 | 1.2099 | 28.88 | 48.89 | 69.7884 |
105
+ | 0.8199 | 0.85 | 2600 | 1.1972 | 29.37 | 48.07 | 67.3120 |
106
+ | 0.6974 | 0.89 | 2700 | 1.1857 | 29.7 | 48.95 | 70.5988 |
107
+ | 0.6736 | 0.92 | 2800 | 1.1884 | 29.33 | 48.97 | 72.7150 |
108
+ | 0.6826 | 0.95 | 2900 | 1.1834 | 30.76 | 50.11 | 68.1225 |
109
+ | 0.7001 | 0.99 | 3000 | 1.1742 | 30.16 | 50.72 | 69.9685 |
110
+
111
+
112
+ ### Framework versions
113
+
114
+ - Transformers 4.39.3
115
+ - Pytorch 2.0.1+cu118
116
+ - Datasets 2.18.0
117
+ - Tokenizers 0.15.2
generation_config.json ADDED
@@ -0,0 +1,261 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 9,
5
+ 19
6
+ ],
7
+ [
8
+ 11,
9
+ 2
10
+ ],
11
+ [
12
+ 11,
13
+ 4
14
+ ],
15
+ [
16
+ 11,
17
+ 17
18
+ ],
19
+ [
20
+ 22,
21
+ 7
22
+ ],
23
+ [
24
+ 22,
25
+ 11
26
+ ],
27
+ [
28
+ 22,
29
+ 17
30
+ ],
31
+ [
32
+ 23,
33
+ 2
34
+ ],
35
+ [
36
+ 23,
37
+ 15
38
+ ]
39
+ ],
40
+ "begin_suppress_tokens": [
41
+ 220,
42
+ 50257
43
+ ],
44
+ "bos_token_id": 50257,
45
+ "decoder_start_token_id": 50258,
46
+ "eos_token_id": 50257,
47
+ "forced_decoder_ids": [
48
+ [
49
+ 1,
50
+ null
51
+ ],
52
+ [
53
+ 2,
54
+ 50359
55
+ ]
56
+ ],
57
+ "is_multilingual": true,
58
+ "lang_to_id": {
59
+ "<|af|>": 50327,
60
+ "<|am|>": 50334,
61
+ "<|ar|>": 50272,
62
+ "<|as|>": 50350,
63
+ "<|az|>": 50304,
64
+ "<|ba|>": 50355,
65
+ "<|be|>": 50330,
66
+ "<|bg|>": 50292,
67
+ "<|bn|>": 50302,
68
+ "<|bo|>": 50347,
69
+ "<|br|>": 50309,
70
+ "<|bs|>": 50315,
71
+ "<|ca|>": 50270,
72
+ "<|cs|>": 50283,
73
+ "<|cy|>": 50297,
74
+ "<|da|>": 50285,
75
+ "<|de|>": 50261,
76
+ "<|el|>": 50281,
77
+ "<|en|>": 50259,
78
+ "<|es|>": 50262,
79
+ "<|et|>": 50307,
80
+ "<|eu|>": 50310,
81
+ "<|fa|>": 50300,
82
+ "<|fi|>": 50277,
83
+ "<|fo|>": 50338,
84
+ "<|fr|>": 50265,
85
+ "<|gl|>": 50319,
86
+ "<|gu|>": 50333,
87
+ "<|haw|>": 50352,
88
+ "<|ha|>": 50354,
89
+ "<|he|>": 50279,
90
+ "<|hi|>": 50276,
91
+ "<|hr|>": 50291,
92
+ "<|ht|>": 50339,
93
+ "<|hu|>": 50286,
94
+ "<|hy|>": 50312,
95
+ "<|id|>": 50275,
96
+ "<|is|>": 50311,
97
+ "<|it|>": 50274,
98
+ "<|ja|>": 50266,
99
+ "<|jw|>": 50356,
100
+ "<|ka|>": 50329,
101
+ "<|kk|>": 50316,
102
+ "<|km|>": 50323,
103
+ "<|kn|>": 50306,
104
+ "<|ko|>": 50264,
105
+ "<|la|>": 50294,
106
+ "<|lb|>": 50345,
107
+ "<|ln|>": 50353,
108
+ "<|lo|>": 50336,
109
+ "<|lt|>": 50293,
110
+ "<|lv|>": 50301,
111
+ "<|mg|>": 50349,
112
+ "<|mi|>": 50295,
113
+ "<|mk|>": 50308,
114
+ "<|ml|>": 50296,
115
+ "<|mn|>": 50314,
116
+ "<|mr|>": 50320,
117
+ "<|ms|>": 50282,
118
+ "<|mt|>": 50343,
119
+ "<|my|>": 50346,
120
+ "<|ne|>": 50313,
121
+ "<|nl|>": 50271,
122
+ "<|nn|>": 50342,
123
+ "<|no|>": 50288,
124
+ "<|oc|>": 50328,
125
+ "<|pa|>": 50321,
126
+ "<|pl|>": 50269,
127
+ "<|ps|>": 50340,
128
+ "<|pt|>": 50267,
129
+ "<|ro|>": 50284,
130
+ "<|ru|>": 50263,
131
+ "<|sa|>": 50344,
132
+ "<|sd|>": 50332,
133
+ "<|si|>": 50322,
134
+ "<|sk|>": 50298,
135
+ "<|sl|>": 50305,
136
+ "<|sn|>": 50324,
137
+ "<|so|>": 50326,
138
+ "<|sq|>": 50317,
139
+ "<|sr|>": 50303,
140
+ "<|su|>": 50357,
141
+ "<|sv|>": 50273,
142
+ "<|sw|>": 50318,
143
+ "<|ta|>": 50287,
144
+ "<|te|>": 50299,
145
+ "<|tg|>": 50331,
146
+ "<|th|>": 50289,
147
+ "<|tk|>": 50341,
148
+ "<|tl|>": 50348,
149
+ "<|tr|>": 50268,
150
+ "<|tt|>": 50351,
151
+ "<|uk|>": 50280,
152
+ "<|ur|>": 50290,
153
+ "<|uz|>": 50337,
154
+ "<|vi|>": 50278,
155
+ "<|yi|>": 50335,
156
+ "<|yo|>": 50325,
157
+ "<|zh|>": 50260
158
+ },
159
+ "language": "en",
160
+ "max_initial_timestamp_index": 50,
161
+ "max_length": 448,
162
+ "no_timestamps_token_id": 50363,
163
+ "pad_token_id": 50257,
164
+ "prev_sot_token_id": 50361,
165
+ "return_timestamps": false,
166
+ "suppress_tokens": [
167
+ 1,
168
+ 2,
169
+ 7,
170
+ 8,
171
+ 9,
172
+ 10,
173
+ 14,
174
+ 25,
175
+ 26,
176
+ 27,
177
+ 28,
178
+ 29,
179
+ 31,
180
+ 58,
181
+ 59,
182
+ 60,
183
+ 61,
184
+ 62,
185
+ 63,
186
+ 90,
187
+ 91,
188
+ 92,
189
+ 93,
190
+ 359,
191
+ 503,
192
+ 522,
193
+ 542,
194
+ 873,
195
+ 893,
196
+ 902,
197
+ 918,
198
+ 922,
199
+ 931,
200
+ 1350,
201
+ 1853,
202
+ 1982,
203
+ 2460,
204
+ 2627,
205
+ 3246,
206
+ 3253,
207
+ 3268,
208
+ 3536,
209
+ 3846,
210
+ 3961,
211
+ 4183,
212
+ 4667,
213
+ 6585,
214
+ 6647,
215
+ 7273,
216
+ 9061,
217
+ 9383,
218
+ 10428,
219
+ 10929,
220
+ 11938,
221
+ 12033,
222
+ 12331,
223
+ 12562,
224
+ 13793,
225
+ 14157,
226
+ 14635,
227
+ 15265,
228
+ 15618,
229
+ 16553,
230
+ 16604,
231
+ 18362,
232
+ 18956,
233
+ 20075,
234
+ 21675,
235
+ 22520,
236
+ 26130,
237
+ 26161,
238
+ 26435,
239
+ 28279,
240
+ 29464,
241
+ 31650,
242
+ 32302,
243
+ 32470,
244
+ 36865,
245
+ 42863,
246
+ 47425,
247
+ 49870,
248
+ 50254,
249
+ 50258,
250
+ 50358,
251
+ 50359,
252
+ 50360,
253
+ 50361,
254
+ 50362
255
+ ],
256
+ "task_to_id": {
257
+ "transcribe": 50359,
258
+ "translate": 50358
259
+ },
260
+ "transformers_version": "4.39.3"
261
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a16db8c3561c99c29ac87f228943df5fb8e70eb7fbb82ce22fd6fe1e218259c2
3
+ size 4992706480
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91b24688e7490ad94ce577f1280d97b405ff4d9d76de4a656b029f2fdec97854
3
+ size 1180663192
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff