Hugging Face's logo

jethrowang
/
vanilla-whisper-tiny

TensorBoard
Safetensors
Chinese
whisper
Generated from Trainer
Model card Files
xet
Metrics Community
jethrowang commited on
Commit
ff21e54
·
verified ·
1 Parent(s): 637c610

End of training

Browse files
Files changed (1) hide show
  1. README.md +13 -13
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-tiny](https://huggingface.co/openai/whisper-tiny) on the HAT ASR Aligned dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.1374
23
- - Cer: 8.7211
24
 
25
  ## Model description
26
 
@@ -39,7 +39,7 @@ More information needed
39
  ### Training hyperparameters
40
 
41
  The following hyperparameters were used during training:
42
- - learning_rate: 0.0001
43
  - train_batch_size: 64
44
  - eval_batch_size: 32
45
  - seed: 42
@@ -53,16 +53,16 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Cer |
55
  |:-------------:|:------:|:-----:|:---------------:|:-------:|
56
- | 0.1 | 0.9993 | 1521 | 0.2944 | 26.2871 |
57
- | 0.0528 | 1.9987 | 3042 | 0.2299 | 26.2362 |
58
- | 0.0318 | 2.9980 | 4563 | 0.2016 | 14.8947 |
59
- | 0.017 | 3.9974 | 6084 | 0.1846 | 17.2273 |
60
- | 0.0093 | 4.9967 | 7605 | 0.1898 | 12.3679 |
61
- | 0.0058 | 5.9961 | 9126 | 0.1629 | 11.8339 |
62
- | 0.0026 | 6.9954 | 10647 | 0.1574 | 9.6620 |
63
- | 0.0016 | 7.9947 | 12168 | 0.1458 | 8.9962 |
64
- | 0.0008 | 8.9941 | 13689 | 0.1404 | 9.7452 |
65
- | 0.0004 | 9.9934 | 15210 | 0.1374 | 8.7211 |
66
 
67
 
68
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the HAT ASR Aligned dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.1729
23
+ - Cer: 10.2307
24
 
25
  ## Model description
26
 
 
39
  ### Training hyperparameters
40
 
41
  The following hyperparameters were used during training:
42
+ - learning_rate: 1e-05
43
  - train_batch_size: 64
44
  - eval_batch_size: 32
45
  - seed: 42
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Cer |
55
  |:-------------:|:------:|:-----:|:---------------:|:-------:|
56
+ | 0.2476 | 0.9993 | 1521 | 0.4437 | 23.6551 |
57
+ | 0.0892 | 1.9987 | 3042 | 0.2482 | 14.6693 |
58
+ | 0.0543 | 2.9980 | 4563 | 0.2007 | 11.1774 |
59
+ | 0.0361 | 3.9974 | 6084 | 0.1847 | 12.4939 |
60
+ | 0.0235 | 4.9967 | 7605 | 0.1791 | 10.5405 |
61
+ | 0.0157 | 5.9961 | 9126 | 0.1727 | 10.9000 |
62
+ | 0.0121 | 6.9954 | 10647 | 0.1724 | 11.1554 |
63
+ | 0.0082 | 7.9947 | 12168 | 0.1720 | 10.3694 |
64
+ | 0.0059 | 8.9941 | 13689 | 0.1732 | 10.4053 |
65
+ | 0.0049 | 9.9934 | 15210 | 0.1729 | 10.2307 |
66
 
67
 
68
  ### Framework versions