Hugging Face's logo

Hanhpt23
/
whisper-small-vietmed-free_E0-8

Automatic Speech Recognition
Transformers
Safetensors
Vietnamese
whisper
Generated from Trainer
Inference Endpoints
Model card Files Community
Hanhpt23 commited on
Commit
e6834ff
•
1 Parent(s): 5bf276b

End of training

Browse files
Files changed (1) hide show
  1. README.md +25 -24
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/VietMed-split-8-2 dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.9095
23
- - Wer: 20.9152
 
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
- | 0.6329 | 1.0 | 569 | 0.6361 | 28.8816 |
56
- | 0.4256 | 2.0 | 1138 | 0.5823 | 25.3926 |
57
- | 0.2837 | 3.0 | 1707 | 0.6103 | 23.9905 |
58
- | 0.159 | 4.0 | 2276 | 0.6727 | 24.7776 |
59
- | 0.0957 | 5.0 | 2845 | 0.7264 | 24.3090 |
60
- | 0.0741 | 6.0 | 3414 | 0.7627 | 23.8733 |
61
- | 0.0427 | 7.0 | 3983 | 0.8138 | 24.0088 |
62
- | 0.0355 | 8.0 | 4552 | 0.8271 | 24.0564 |
63
- | 0.0325 | 9.0 | 5121 | 0.8324 | 23.7196 |
64
- | 0.0182 | 10.0 | 5690 | 0.8472 | 23.1741 |
65
- | 0.0107 | 11.0 | 6259 | 0.8849 | 22.2625 |
66
- | 0.0095 | 12.0 | 6828 | 0.8820 | 22.3211 |
67
- | 0.0045 | 13.0 | 7397 | 0.8804 | 21.9403 |
68
- | 0.0039 | 14.0 | 7966 | 0.8763 | 21.8708 |
69
- | 0.0036 | 15.0 | 8535 | 0.8695 | 22.0099 |
70
- | 0.0016 | 16.0 | 9104 | 0.8850 | 21.4351 |
71
- | 0.0005 | 17.0 | 9673 | 0.9032 | 21.0214 |
72
- | 0.0004 | 18.0 | 10242 | 0.8993 | 21.1788 |
73
- | 0.0001 | 19.0 | 10811 | 0.9081 | 20.9006 |
74
- | 0.0002 | 20.0 | 11380 | 0.9095 | 20.9152 |
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/VietMed-split-8-2 dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.9214
23
+ - Wer: 21.2777
24
+ - Cer: 17.6019
25
 
26
  ## Model description
27
 
 
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
55
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
56
+ | 0.624 | 1.0 | 569 | 0.6172 | 28.3507 | 24.2846 |
57
+ | 0.4335 | 2.0 | 1138 | 0.5851 | 24.6604 | 20.0193 |
58
+ | 0.2821 | 3.0 | 1707 | 0.6146 | 24.1333 | 19.6217 |
59
+ | 0.1837 | 4.0 | 2276 | 0.6578 | 24.6458 | 20.3639 |
60
+ | 0.0938 | 5.0 | 2845 | 0.7273 | 24.2724 | 19.7762 |
61
+ | 0.0787 | 6.0 | 3414 | 0.7621 | 23.7232 | 19.1921 |
62
+ | 0.0536 | 7.0 | 3983 | 0.8131 | 23.9649 | 19.5506 |
63
+ | 0.0424 | 8.0 | 4552 | 0.8392 | 24.1552 | 19.7242 |
64
+ | 0.0205 | 9.0 | 5121 | 0.8504 | 22.8409 | 18.6722 |
65
+ | 0.018 | 10.0 | 5690 | 0.8700 | 23.0240 | 18.7208 |
66
+ | 0.016 | 11.0 | 6259 | 0.8755 | 22.8666 | 18.6357 |
67
+ | 0.0152 | 12.0 | 6828 | 0.8713 | 22.5773 | 18.3892 |
68
+ | 0.0062 | 13.0 | 7397 | 0.8858 | 21.9623 | 18.0203 |
69
+ | 0.0051 | 14.0 | 7966 | 0.9012 | 21.9001 | 18.0021 |
70
+ | 0.0024 | 15.0 | 8535 | 0.8900 | 21.7133 | 17.8484 |
71
+ | 0.001 | 16.0 | 9104 | 0.9155 | 21.9586 | 18.0594 |
72
+ | 0.0006 | 17.0 | 9673 | 0.9122 | 21.5047 | 17.7451 |
73
+ | 0.0005 | 18.0 | 10242 | 0.9147 | 21.2997 | 17.5880 |
74
+ | 0.0011 | 19.0 | 10811 | 0.9194 | 21.3143 | 17.6132 |
75
+ | 0.0001 | 20.0 | 11380 | 0.9214 | 21.2777 | 17.6019 |
76
 
77
 
78
  ### Framework versions