ymoslem commited on
Commit
1860ac9
1 Parent(s): 63f395e

End of training

Browse files
Files changed (2) hide show
  1. README.md +88 -4
  2. model.safetensors +1 -1
README.md CHANGED
@@ -1,19 +1,47 @@
1
  ---
 
 
 
2
  license: apache-2.0
3
  base_model: openai/whisper-small
4
  tags:
5
  - generated_from_trainer
 
 
 
 
 
 
 
6
  model-index:
7
- - name: whisper-small-ga2en-v1.3
8
- results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  ---
10
 
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
  should probably proofread and complete it, then remove this comment. -->
13
 
14
- # whisper-small-ga2en-v1.3
15
 
16
- This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
 
 
 
 
 
17
 
18
  ## Model description
19
 
@@ -42,6 +70,62 @@ The following hyperparameters were used during training:
42
  - training_steps: 5000
43
  - mixed_precision_training: Native AMP
44
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45
  ### Framework versions
46
 
47
  - Transformers 4.39.3
 
1
  ---
2
+ language:
3
+ - ga
4
+ - en
5
  license: apache-2.0
6
  base_model: openai/whisper-small
7
  tags:
8
  - generated_from_trainer
9
+ datasets:
10
+ - ymoslem/IWSLT2023-GA-EN
11
+ - ymoslem/FLEURS-GA-EN
12
+ - ymoslem/BitesizeIrish-GA-EN
13
+ metrics:
14
+ - bleu
15
+ - wer
16
  model-index:
17
+ - name: Whisper Small GA-EN Speech Translation, fine-tuned from 1.2, without SpokenWords
18
+ results:
19
+ - task:
20
+ name: Automatic Speech Recognition
21
+ type: automatic-speech-recognition
22
+ dataset:
23
+ name: IWSLT-2023, FLEURS, and BiteSize
24
+ type: ymoslem/IWSLT2023-GA-EN
25
+ metrics:
26
+ - name: Bleu
27
+ type: bleu
28
+ value: 29.96
29
+ - name: Wer
30
+ type: wer
31
+ value: 66.9968482665466
32
  ---
33
 
34
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
35
  should probably proofread and complete it, then remove this comment. -->
36
 
37
+ # Whisper Small GA-EN Speech Translation, fine-tuned from 1.2, without SpokenWords
38
 
39
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the IWSLT-2023, FLEURS, and BiteSize dataset.
40
+ It achieves the following results on the evaluation set:
41
+ - Loss: 1.7177
42
+ - Bleu: 29.96
43
+ - Chrf: 45.61
44
+ - Wer: 66.9968
45
 
46
  ## Model description
47
 
 
70
  - training_steps: 5000
71
  - mixed_precision_training: Native AMP
72
 
73
+ ### Training results
74
+
75
+ | Training Loss | Epoch | Step | Bleu | Chrf | Validation Loss | Wer |
76
+ |:-------------:|:-----:|:----:|:-----:|:-----:|:---------------:|:--------:|
77
+ | 2.4954 | 0.11 | 100 | 3.7 | 18.03 | 2.1286 | 179.7839 |
78
+ | 2.045 | 0.22 | 200 | 12.65 | 25.53 | 1.8146 | 100.9005 |
79
+ | 1.7928 | 0.32 | 300 | 13.78 | 30.2 | 1.7253 | 101.9811 |
80
+ | 1.6615 | 0.43 | 400 | 15.8 | 31.88 | 1.6834 | 92.5259 |
81
+ | 1.4491 | 0.54 | 500 | 15.61 | 36.27 | 1.5971 | 107.3841 |
82
+ | 1.2074 | 0.65 | 600 | 19.92 | 36.31 | 1.5939 | 84.3314 |
83
+ | 1.2308 | 0.76 | 700 | 20.37 | 38.72 | 1.5234 | 84.8267 |
84
+ | 1.107 | 0.86 | 800 | 21.35 | 37.87 | 1.5460 | 82.8906 |
85
+ | 0.9491 | 0.97 | 900 | 21.06 | 40.74 | 1.5161 | 82.5754 |
86
+ | 0.384 | 1.08 | 1000 | 23.24 | 41.98 | 1.4927 | 82.2152 |
87
+ | 0.362 | 1.19 | 1100 | 23.19 | 42.24 | 1.5567 | 80.2792 |
88
+ | 0.3756 | 1.29 | 1200 | 27.83 | 43.8 | 1.5265 | 69.2481 |
89
+ | 0.3401 | 1.4 | 1300 | 21.79 | 41.66 | 1.5522 | 92.3908 |
90
+ | 0.3346 | 1.51 | 1400 | 24.61 | 42.15 | 1.5085 | 75.4615 |
91
+ | 0.3101 | 1.62 | 1500 | 26.67 | 43.41 | 1.4933 | 70.7789 |
92
+ | 0.3231 | 1.73 | 1600 | 27.95 | 42.82 | 1.4979 | 68.3026 |
93
+ | 0.2665 | 1.83 | 1700 | 28.5 | 43.76 | 1.4977 | 68.1225 |
94
+ | 0.2704 | 1.94 | 1800 | 28.15 | 43.87 | 1.5063 | 68.8429 |
95
+ | 0.0769 | 2.05 | 1900 | 25.76 | 43.22 | 1.5162 | 77.6227 |
96
+ | 0.0597 | 2.16 | 2000 | 25.04 | 43.15 | 1.5216 | 79.0635 |
97
+ | 0.0743 | 2.27 | 2100 | 27.85 | 44.43 | 1.5313 | 68.3926 |
98
+ | 0.0878 | 2.37 | 2200 | 27.54 | 43.96 | 1.5495 | 68.3476 |
99
+ | 0.0712 | 2.48 | 2300 | 28.28 | 44.39 | 1.5355 | 65.8712 |
100
+ | 0.0789 | 2.59 | 2400 | 28.64 | 44.75 | 1.5277 | 65.7812 |
101
+ | 0.073 | 2.7 | 2500 | 29.09 | 44.65 | 1.5327 | 65.7812 |
102
+ | 0.073 | 2.8 | 2600 | 25.26 | 43.44 | 1.5304 | 78.2981 |
103
+ | 0.0697 | 2.91 | 2700 | 25.71 | 43.02 | 1.5460 | 78.4782 |
104
+ | 0.0398 | 3.02 | 2800 | 28.26 | 44.71 | 1.5580 | 72.8501 |
105
+ | 0.0302 | 3.13 | 2900 | 30.25 | 45.46 | 1.5688 | 66.1414 |
106
+ | 0.0424 | 3.24 | 3000 | 29.88 | 45.21 | 1.5693 | 66.0964 |
107
+ | 0.0397 | 3.34 | 3100 | 30.01 | 45.85 | 1.5934 | 65.6911 |
108
+ | 0.0346 | 3.45 | 3200 | 30.2 | 45.8 | 1.5818 | 65.8262 |
109
+ | 0.032 | 3.56 | 3300 | 29.81 | 46.5 | 1.5823 | 66.7267 |
110
+ | 0.0348 | 3.67 | 3400 | 30.77 | 46.43 | 1.5752 | 64.6556 |
111
+ | 0.0522 | 5.97 | 3500 | 1.6080| 29.69 | 45.47 | 65.8712 |
112
+ | 0.0443 | 6.14 | 3600 | 1.6272| 29.54 | 44.71 | 65.1508 |
113
+ | 0.0492 | 6.31 | 3700 | 1.6211| 29.3 | 45.36 | 68.3926 |
114
+ | 0.0544 | 6.48 | 3800 | 1.6069| 30.08 | 44.39 | 64.9257 |
115
+ | 0.0574 | 6.66 | 3900 | 1.6306| 28.86 | 44.6 | 66.2765 |
116
+ | 0.0535 | 6.83 | 4000 | 1.6722| 27.92 | 43.48 | 67.9874 |
117
+ | 0.0424 | 7.0 | 4100 | 1.6968| 27.48 | 44.29 | 70.3737 |
118
+ | 0.0235 | 7.17 | 4200 | 1.6768| 27.97 | 45.34 | 70.0135 |
119
+ | 0.0262 | 7.34 | 4300 | 1.6908| 28.77 | 45.74 | 68.3926 |
120
+ | 0.0218 | 7.51 | 4400 | 1.6890| 28.97 | 46.57 | 69.5182 |
121
+ | 0.0293 | 7.68 | 4500 | 1.6742| 29.51 | 45.38 | 68.8429 |
122
+ | 0.0194 | 7.85 | 4600 | 1.6962| 29.63 | 45.18 | 67.9874 |
123
+ | 0.0187 | 8.02 | 4700 | 1.6936| 30.1 | 45.28 | 66.0964 |
124
+ | 0.0115 | 8.19 | 4800 | 1.7162| 30.0 | 46.02 | 67.6722 |
125
+ | 0.0138 | 8.36 | 4900 | 1.7113| 30.34 | 46.01 | 66.6817 |
126
+ | 0.0098 | 8.53 | 5000 | 1.7177| 29.96 | 45.61 | 66.9968 |
127
+
128
+
129
  ### Framework versions
130
 
131
  - Transformers 4.39.3
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:469683146728fb5e4d38e825310a71de263c3c2701ae3841124733d01fc69f7a
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1f5c86cf3c030450c76958f86a536f891dd178ae53c000827c0a14969981a61
3
  size 966995080