Hugging Face's logo

GleamEyeBeast
/
ascend

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Model card Files
xet
Metrics Community
1
GleamEyeBeast commited on
Commit
52c68be
·
1 Parent(s): 83fa864

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -16
README.md CHANGED
@@ -13,9 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  This model is a fine-tuned version of [GleamEyeBeast/ascend](https://huggingface.co/GleamEyeBeast/ascend) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
- - Loss: 1.1048
17
- - Wer: 0.7863
18
- - Cer: 0.3097
19
 
20
  ## Model description
21
 
@@ -40,23 +40,33 @@ The following hyperparameters were used during training:
40
  - seed: 42
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
43
- - num_epochs: 10
44
  - mixed_precision_training: Native AMP
45
 
46
  ### Training results
47
 
48
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
49
- |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
50
- | 2.1836 | 1.0 | 688 | 1.4966 | 0.8833 | 0.5053 |
51
- | 1.4799 | 2.0 | 1376 | 1.2656 | 0.8818 | 0.4452 |
52
- | 1.0552 | 3.0 | 2064 | 1.2256 | 0.9278 | 0.3741 |
53
- | 0.9307 | 4.0 | 2752 | 1.1689 | 0.8315 | 0.3527 |
54
- | 0.8515 | 5.0 | 3440 | 1.1272 | 0.8078 | 0.3452 |
55
- | 0.7176 | 6.0 | 4128 | 1.1201 | 0.8351 | 0.3280 |
56
- | 0.6819 | 7.0 | 4816 | 1.1073 | 0.7999 | 0.3230 |
57
- | 0.6103 | 8.0 | 5504 | 1.1128 | 0.7956 | 0.3153 |
58
- | 0.5741 | 9.0 | 6192 | 1.1113 | 0.7859 | 0.3105 |
59
- | 0.5645 | 10.0 | 6880 | 1.1048 | 0.7863 | 0.3097 |
 
 
 
 
 
 
 
 
 
 
60
 
61
 
62
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [GleamEyeBeast/ascend](https://huggingface.co/GleamEyeBeast/ascend) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Loss: 1.3718
17
+ - Wer: 0.6412
18
+ - Cer: 0.2428
19
 
20
  ## Model description
21
 
 
40
  - seed: 42
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
43
+ - num_epochs: 20
44
  - mixed_precision_training: Native AMP
45
 
46
  ### Training results
47
 
48
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
49
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
50
+ | 0.5769 | 1.0 | 688 | 1.1864 | 0.7716 | 0.3159 |
51
+ | 0.5215 | 2.0 | 1376 | 1.1613 | 0.7504 | 0.2965 |
52
+ | 0.4188 | 3.0 | 2064 | 1.1644 | 0.7389 | 0.2950 |
53
+ | 0.3695 | 4.0 | 2752 | 1.1937 | 0.7184 | 0.2815 |
54
+ | 0.3404 | 5.0 | 3440 | 1.1947 | 0.7083 | 0.2719 |
55
+ | 0.2885 | 6.0 | 4128 | 1.2314 | 0.7108 | 0.2685 |
56
+ | 0.2727 | 7.0 | 4816 | 1.2243 | 0.6850 | 0.2616 |
57
+ | 0.2417 | 8.0 | 5504 | 1.2506 | 0.6767 | 0.2608 |
58
+ | 0.2207 | 9.0 | 6192 | 1.2804 | 0.6922 | 0.2595 |
59
+ | 0.2195 | 10.0 | 6880 | 1.2582 | 0.6818 | 0.2575 |
60
+ | 0.1896 | 11.0 | 7568 | 1.3101 | 0.6814 | 0.2545 |
61
+ | 0.1961 | 12.0 | 8256 | 1.2793 | 0.6706 | 0.2526 |
62
+ | 0.1752 | 13.0 | 8944 | 1.2643 | 0.6584 | 0.2509 |
63
+ | 0.1638 | 14.0 | 9632 | 1.3152 | 0.6588 | 0.2482 |
64
+ | 0.1522 | 15.0 | 10320 | 1.3098 | 0.6433 | 0.2439 |
65
+ | 0.1351 | 16.0 | 11008 | 1.3253 | 0.6537 | 0.2447 |
66
+ | 0.1266 | 17.0 | 11696 | 1.3394 | 0.6365 | 0.2418 |
67
+ | 0.1289 | 18.0 | 12384 | 1.3718 | 0.6412 | 0.2443 |
68
+ | 0.1204 | 19.0 | 13072 | 1.3708 | 0.6433 | 0.2433 |
69
+ | 0.1189 | 20.0 | 13760 | 1.3718 | 0.6412 | 0.2428 |
70
 
71
 
72
  ### Framework versions