Hugging Face's logo

Hanhpt23
/
whisper-small-frenchmed-free_E0-8

Automatic Speech Recognition
Transformers
Safetensors
French
whisper
Generated from Trainer
Inference Endpoints
Model card Files Community
Hanhpt23 commited on
Commit
c0c6f72
•
1 Parent(s): 2cdc963

End of training

Browse files
Files changed (2) hide show
  1. README.md +25 -24
  2. model.safetensors +1 -1
README.md CHANGED
@@ -19,8 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the pphuc25/FrenchMed dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 1.4773
23
- - Wer: 36.7302
 
24
 
25
  ## Model description
26
 
@@ -50,28 +51,28 @@ The following hyperparameters were used during training:
50
 
51
  ### Training results
52
 
53
- | Training Loss | Epoch | Step | Validation Loss | Wer |
54
- |:-------------:|:-----:|:----:|:---------------:|:-------:|
55
- | 1.0463 | 1.0 | 215 | 1.0179 | 55.0587 |
56
- | 0.6082 | 2.0 | 430 | 1.0745 | 57.8446 |
57
- | 0.3176 | 3.0 | 645 | 1.1829 | 41.1290 |
58
- | 0.1609 | 4.0 | 860 | 1.2342 | 41.7155 |
59
- | 0.1106 | 5.0 | 1075 | 1.2716 | 39.7361 |
60
- | 0.0895 | 6.0 | 1290 | 1.3019 | 40.3226 |
61
- | 0.0761 | 7.0 | 1505 | 1.3814 | 42.0088 |
62
- | 0.056 | 8.0 | 1720 | 1.4171 | 42.3754 |
63
- | 0.0379 | 9.0 | 1935 | 1.4578 | 40.6891 |
64
- | 0.0278 | 10.0 | 2150 | 1.4600 | 40.2493 |
65
- | 0.0235 | 11.0 | 2365 | 1.4685 | 37.9765 |
66
- | 0.0134 | 12.0 | 2580 | 1.4823 | 40.1760 |
67
- | 0.0129 | 13.0 | 2795 | 1.4950 | 39.4428 |
68
- | 0.008 | 14.0 | 3010 | 1.4921 | 39.0762 |
69
- | 0.0038 | 15.0 | 3225 | 1.4791 | 36.0704 |
70
- | 0.0042 | 16.0 | 3440 | 1.4517 | 39.4428 |
71
- | 0.0025 | 17.0 | 3655 | 1.4607 | 37.9032 |
72
- | 0.0006 | 18.0 | 3870 | 1.4702 | 36.2903 |
73
- | 0.0002 | 19.0 | 4085 | 1.4753 | 36.6569 |
74
- | 0.0006 | 20.0 | 4300 | 1.4773 | 36.7302 |
75
 
76
 
77
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the pphuc25/FrenchMed dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 1.4691
23
+ - Wer: 40.9824
24
+ - Cer: 28.8290
25
 
26
  ## Model description
27
 
 
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
55
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
56
+ | 1.054 | 1.0 | 215 | 1.0167 | 52.2727 | 33.9067 |
57
+ | 0.6007 | 2.0 | 430 | 1.0289 | 57.0381 | 35.2828 |
58
+ | 0.3026 | 3.0 | 645 | 1.1406 | 40.3226 | 27.7418 |
59
+ | 0.1617 | 4.0 | 860 | 1.2025 | 40.0293 | 25.8291 |
60
+ | 0.1133 | 5.0 | 1075 | 1.2893 | 42.0088 | 28.0583 |
61
+ | 0.0903 | 6.0 | 1290 | 1.2941 | 42.0088 | 29.6821 |
62
+ | 0.0821 | 7.0 | 1505 | 1.3255 | 41.5689 | 28.0033 |
63
+ | 0.0511 | 8.0 | 1720 | 1.4011 | 43.5484 | 28.2923 |
64
+ | 0.0391 | 9.0 | 1935 | 1.4531 | 39.3695 | 27.4804 |
65
+ | 0.0283 | 10.0 | 2150 | 1.4935 | 43.0352 | 29.9436 |
66
+ | 0.0181 | 11.0 | 2365 | 1.4941 | 43.0352 | 29.1042 |
67
+ | 0.0152 | 12.0 | 2580 | 1.4768 | 43.3284 | 31.3609 |
68
+ | 0.0191 | 13.0 | 2795 | 1.4495 | 43.9150 | 30.8380 |
69
+ | 0.0055 | 14.0 | 3010 | 1.4699 | 42.1554 | 29.2418 |
70
+ | 0.0032 | 15.0 | 3225 | 1.4493 | 39.8094 | 28.0859 |
71
+ | 0.0041 | 16.0 | 3440 | 1.4503 | 40.8358 | 28.7739 |
72
+ | 0.0018 | 17.0 | 3655 | 1.4598 | 41.6422 | 29.3931 |
73
+ | 0.0015 | 18.0 | 3870 | 1.4654 | 41.2023 | 28.7051 |
74
+ | 0.0002 | 19.0 | 4085 | 1.4676 | 40.8358 | 28.7051 |
75
+ | 0.0005 | 20.0 | 4300 | 1.4691 | 40.9824 | 28.8290 |
76
 
77
 
78
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:21fb10d75d0188878033b375786c0b5bfa3bf330288396e20f97fc5f6e230f2f
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8215d17de63662b6a9504b9690187c9bab76962555e05710d8597db2d1f57ca8
3
  size 966995080