Hugging Face's logo

Hanhpt23
/
whisper-small-vietmed-free_E3-11

Automatic Speech Recognition
Transformers
Safetensors
Vietnamese
whisper
Generated from Trainer
Inference Endpoints
Model card Files Community
Hanhpt23 commited on
Commit
1069d1e
•
1 Parent(s): 620e3f4

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: 0.9118
23
- - Wer: 21.7024
 
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
- | 0.6339 | 1.0 | 569 | 0.6214 | 27.8675 |
56
- | 0.4322 | 2.0 | 1138 | 0.5860 | 25.1949 |
57
- | 0.2846 | 3.0 | 1707 | 0.6092 | 24.0564 |
58
- | 0.1624 | 4.0 | 2276 | 0.6697 | 24.2614 |
59
- | 0.0915 | 5.0 | 2845 | 0.7291 | 25.9162 |
60
- | 0.0642 | 6.0 | 3414 | 0.7756 | 24.7739 |
61
- | 0.0382 | 7.0 | 3983 | 0.8095 | 24.1955 |
62
- | 0.0355 | 8.0 | 4552 | 0.8362 | 24.2541 |
63
- | 0.0314 | 9.0 | 5121 | 0.8388 | 23.9063 |
64
- | 0.0174 | 10.0 | 5690 | 0.8719 | 22.9251 |
65
- | 0.0107 | 11.0 | 6259 | 0.8740 | 23.0899 |
66
- | 0.0109 | 12.0 | 6828 | 0.8811 | 22.8226 |
67
- | 0.0036 | 13.0 | 7397 | 0.8770 | 22.5700 |
68
- | 0.0037 | 14.0 | 7966 | 0.8888 | 22.4382 |
69
- | 0.002 | 15.0 | 8535 | 0.8917 | 22.3467 |
70
- | 0.0001 | 16.0 | 9104 | 0.9047 | 22.0099 |
71
- | 0.0004 | 17.0 | 9673 | 0.8960 | 21.9660 |
72
- | 0.0009 | 18.0 | 10242 | 0.9079 | 21.6914 |
73
- | 0.0001 | 19.0 | 10811 | 0.9108 | 21.6658 |
74
- | 0.0 | 20.0 | 11380 | 0.9118 | 21.7024 |
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: 0.9074
23
+ - Wer: 21.2813
24
+ - Cer: 17.6054
25
 
26
  ## Model description
27
 
 
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
55
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
56
+ | 0.631 | 1.0 | 569 | 0.6148 | 26.4031 | 20.9932 |
57
+ | 0.4198 | 2.0 | 1138 | 0.5805 | 24.4042 | 19.7276 |
58
+ | 0.2593 | 3.0 | 1707 | 0.6228 | 24.8545 | 20.0158 |
59
+ | 0.1615 | 4.0 | 2276 | 0.6910 | 24.0161 | 19.3631 |
60
+ | 0.086 | 5.0 | 2845 | 0.7390 | 24.2065 | 19.7389 |
61
+ | 0.0612 | 6.0 | 3414 | 0.7867 | 24.3053 | 19.6096 |
62
+ | 0.0467 | 7.0 | 3983 | 0.8099 | 23.6280 | 19.0827 |
63
+ | 0.0366 | 8.0 | 4552 | 0.8577 | 23.9868 | 19.4967 |
64
+ | 0.0245 | 9.0 | 5121 | 0.8748 | 23.6280 | 19.3119 |
65
+ | 0.0166 | 10.0 | 5690 | 0.8653 | 23.1558 | 18.9742 |
66
+ | 0.011 | 11.0 | 6259 | 0.8834 | 23.7452 | 19.4160 |
67
+ | 0.0139 | 12.0 | 6828 | 0.8843 | 23.3571 | 19.2424 |
68
+ | 0.0038 | 13.0 | 7397 | 0.8823 | 22.1600 | 18.0194 |
69
+ | 0.0097 | 14.0 | 7966 | 0.8805 | 22.5334 | 18.3371 |
70
+ | 0.0022 | 15.0 | 8535 | 0.8891 | 21.7573 | 17.8753 |
71
+ | 0.0017 | 16.0 | 9104 | 0.8898 | 21.8854 | 18.0116 |
72
+ | 0.0002 | 17.0 | 9673 | 0.8969 | 21.5047 | 17.7807 |
73
+ | 0.0016 | 18.0 | 10242 | 0.9140 | 21.3033 | 17.5880 |
74
+ | 0.0001 | 19.0 | 10811 | 0.9044 | 21.3070 | 17.6054 |
75
+ | 0.0001 | 20.0 | 11380 | 0.9074 | 21.2813 | 17.6054 |
76
 
77
 
78
  ### Framework versions