Hugging Face's logo

Hanhpt23
/
whisper-small-engmed-v1

Automatic Speech Recognition
Transformers
Safetensors
English
whisper
Generated from Trainer
Inference Endpoints
Model card Files Community
Hanhpt23 commited on
Commit
3f0b4e4
•
1 Parent(s): 1c1cbd3

End of training

Browse files
Files changed (2) hide show
  1. README.md +24 -24
  2. model.safetensors +1 -1
README.md CHANGED
@@ -19,8 +19,8 @@ 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/EngMed dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 3.0716
23
- - Wer: 52.2789
24
 
25
  ## Model description
26
 
@@ -50,28 +50,28 @@ The following hyperparameters were used during training:
50
 
51
  ### Training results
52
 
53
- | Training Loss | Epoch | Step | Validation Loss | Wer |
54
- |:-------------:|:-----:|:-----:|:---------------:|:-------:|
55
- | 1.1558 | 1.0 | 2506 | 1.7618 | 61.5410 |
56
- | 0.7056 | 2.0 | 5012 | 1.8179 | 51.1240 |
57
- | 0.4179 | 3.0 | 7518 | 2.0105 | 51.7298 |
58
- | 0.2239 | 4.0 | 10024 | 2.2419 | 55.2747 |
59
- | 0.1215 | 5.0 | 12530 | 2.4194 | 52.5344 |
60
- | 0.1093 | 6.0 | 15036 | 2.5205 | 52.8089 |
61
- | 0.0801 | 7.0 | 17542 | 2.6161 | 55.1091 |
62
- | 0.0546 | 8.0 | 20048 | 2.6225 | 54.4370 |
63
- | 0.0434 | 9.0 | 22554 | 2.7782 | 53.2207 |
64
- | 0.0259 | 10.0 | 25060 | 2.8240 | 54.5459 |
65
- | 0.0261 | 11.0 | 27566 | 2.8877 | 53.2349 |
66
- | 0.0082 | 12.0 | 30072 | 2.9535 | 51.8056 |
67
- | 0.0086 | 13.0 | 32578 | 2.9738 | 53.7886 |
68
- | 0.0206 | 14.0 | 35084 | 2.9812 | 52.5818 |
69
- | 0.0036 | 15.0 | 37590 | 2.9906 | 53.1213 |
70
- | 0.0028 | 16.0 | 40096 | 2.9971 | 52.7616 |
71
- | 0.0048 | 17.0 | 42602 | 3.0191 | 52.6811 |
72
- | 0.0048 | 18.0 | 45108 | 3.0563 | 51.2660 |
73
- | 0.0029 | 19.0 | 47614 | 3.0625 | 51.9381 |
74
- | 0.001 | 20.0 | 50120 | 3.0716 | 52.2789 |
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/EngMed dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 1.1935
23
+ - Wer: 27.2679
24
 
25
  ## Model description
26
 
 
50
 
51
  ### Training results
52
 
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|
55
+ | 0.5346 | 1.0 | 323 | 0.6308 | 30.4852 |
56
+ | 0.2978 | 2.0 | 646 | 0.7524 | 25.0527 |
57
+ | 0.1478 | 3.0 | 969 | 0.7934 | 29.2194 |
58
+ | 0.1208 | 4.0 | 1292 | 0.8305 | 34.0190 |
59
+ | 0.0667 | 5.0 | 1615 | 0.8781 | 22.2046 |
60
+ | 0.0562 | 6.0 | 1938 | 1.0234 | 27.0042 |
61
+ | 0.0519 | 7.0 | 2261 | 0.9954 | 17.1941 |
62
+ | 0.0311 | 8.0 | 2584 | 1.0153 | 35.2321 |
63
+ | 0.0183 | 9.0 | 2907 | 1.1135 | 48.5759 |
64
+ | 0.015 | 10.0 | 3230 | 1.1428 | 35.6540 |
65
+ | 0.0099 | 11.0 | 3553 | 1.1355 | 39.3987 |
66
+ | 0.0054 | 12.0 | 3876 | 1.1408 | 27.3207 |
67
+ | 0.0037 | 13.0 | 4199 | 1.1651 | 28.6920 |
68
+ | 0.0024 | 14.0 | 4522 | 1.1648 | 40.2954 |
69
+ | 0.0004 | 15.0 | 4845 | 1.1480 | 44.0401 |
70
+ | 0.0004 | 16.0 | 5168 | 1.1564 | 23.1540 |
71
+ | 0.0003 | 17.0 | 5491 | 1.1754 | 28.2173 |
72
+ | 0.0007 | 18.0 | 5814 | 1.1884 | 26.0021 |
73
+ | 0.0001 | 19.0 | 6137 | 1.1925 | 27.9536 |
74
+ | 0.0001 | 20.0 | 6460 | 1.1935 | 27.2679 |
75
 
76
 
77
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d807aceb82c26ff71aa18bb84560f599e5652ee2f91148303932cabbdec4aad9
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1574878897a182c75da98d1d931a3db4cb7198e6e959b0a62b3485a81160c578
3
  size 966995080