yip-i commited on
Commit
c1a7570
·
1 Parent(s): 1df0b34

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +35 -35
README.md CHANGED
@@ -14,8 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.3092
18
- - Wer: 0.4970
19
 
20
  ## Model description
21
 
@@ -47,39 +47,39 @@ The following hyperparameters were used during training:
47
 
48
  | Training Loss | Epoch | Step | Validation Loss | Wer |
49
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
50
- | 23.1195 | 0.89 | 500 | 3.3626 | 1.0 |
51
- | 3.2757 | 1.79 | 1000 | 3.0206 | 1.0 |
52
- | 2.9154 | 2.68 | 1500 | 2.8301 | 1.0 |
53
- | 2.6907 | 3.57 | 2000 | 2.3674 | 1.0803 |
54
- | 2.005 | 4.46 | 2500 | 1.2827 | 1.2962 |
55
- | 1.3517 | 5.36 | 3000 | 0.7531 | 1.1058 |
56
- | 1.0444 | 6.25 | 3500 | 0.6424 | 1.0079 |
57
- | 0.8084 | 7.14 | 4000 | 0.4733 | 0.8729 |
58
- | 0.6616 | 8.04 | 4500 | 0.4022 | 0.8078 |
59
- | 0.5762 | 8.93 | 5000 | 0.3708 | 0.7828 |
60
- | 0.5198 | 9.82 | 5500 | 0.3335 | 0.7184 |
61
- | 0.443 | 10.71 | 6000 | 0.3298 | 0.7026 |
62
- | 0.4203 | 11.61 | 6500 | 0.3004 | 0.6734 |
63
- | 0.3533 | 12.5 | 7000 | 0.2846 | 0.6600 |
64
- | 0.3448 | 13.39 | 7500 | 0.3020 | 0.6411 |
65
- | 0.3045 | 14.29 | 8000 | 0.2823 | 0.6071 |
66
- | 0.2938 | 15.18 | 8500 | 0.2644 | 0.5827 |
67
- | 0.2728 | 16.07 | 9000 | 0.2853 | 0.5827 |
68
- | 0.2601 | 16.96 | 9500 | 0.2767 | 0.5736 |
69
- | 0.233 | 17.86 | 10000 | 0.3136 | 0.5560 |
70
- | 0.2284 | 18.75 | 10500 | 0.2696 | 0.5262 |
71
- | 0.2027 | 19.64 | 11000 | 0.3001 | 0.5268 |
72
- | 0.1914 | 20.54 | 11500 | 0.2858 | 0.5298 |
73
- | 0.1824 | 21.43 | 12000 | 0.2848 | 0.5164 |
74
- | 0.1903 | 22.32 | 12500 | 0.2894 | 0.5207 |
75
- | 0.1619 | 23.21 | 13000 | 0.2731 | 0.5134 |
76
- | 0.17 | 24.11 | 13500 | 0.2557 | 0.5085 |
77
- | 0.1351 | 25.0 | 14000 | 0.2610 | 0.5049 |
78
- | 0.1592 | 25.89 | 14500 | 0.3017 | 0.4988 |
79
- | 0.1482 | 26.79 | 15000 | 0.3018 | 0.5 |
80
- | 0.1376 | 27.68 | 15500 | 0.2882 | 0.4964 |
81
- | 0.1404 | 28.57 | 16000 | 0.3024 | 0.5 |
82
- | 0.1241 | 29.46 | 16500 | 0.3092 | 0.4970 |
83
 
84
 
85
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.3203
18
+ - Wer: 0.5353
19
 
20
  ## Model description
21
 
 
47
 
48
  | Training Loss | Epoch | Step | Validation Loss | Wer |
49
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
50
+ | 23.5576 | 0.89 | 500 | 3.3654 | 1.0 |
51
+ | 3.3953 | 1.79 | 1000 | 3.1729 | 1.0 |
52
+ | 2.9514 | 2.68 | 1500 | 2.8946 | 1.0 |
53
+ | 2.84 | 3.57 | 2000 | 2.8386 | 1.0 |
54
+ | 2.7685 | 4.46 | 2500 | 2.7147 | 1.0 |
55
+ | 2.5059 | 5.36 | 3000 | 2.1341 | 1.1752 |
56
+ | 1.8907 | 6.25 | 3500 | 1.3604 | 1.2403 |
57
+ | 1.3892 | 7.14 | 4000 | 0.8814 | 1.1989 |
58
+ | 1.0754 | 8.04 | 4500 | 0.6416 | 1.0529 |
59
+ | 0.8795 | 8.93 | 5000 | 0.5760 | 0.9641 |
60
+ | 0.7478 | 9.82 | 5500 | 0.4633 | 0.8790 |
61
+ | 0.6107 | 10.71 | 6000 | 0.3921 | 0.8394 |
62
+ | 0.5445 | 11.61 | 6500 | 0.3579 | 0.7987 |
63
+ | 0.4788 | 12.5 | 7000 | 0.3034 | 0.7470 |
64
+ | 0.4435 | 13.39 | 7500 | 0.2989 | 0.7311 |
65
+ | 0.4057 | 14.29 | 8000 | 0.3366 | 0.7092 |
66
+ | 0.3606 | 15.18 | 8500 | 0.2783 | 0.6892 |
67
+ | 0.343 | 16.07 | 9000 | 0.2593 | 0.6612 |
68
+ | 0.3189 | 16.96 | 9500 | 0.2780 | 0.6460 |
69
+ | 0.277 | 17.86 | 10000 | 0.3266 | 0.6277 |
70
+ | 0.2789 | 18.75 | 10500 | 0.3582 | 0.6253 |
71
+ | 0.2552 | 19.64 | 11000 | 0.3422 | 0.6156 |
72
+ | 0.2416 | 20.54 | 11500 | 0.3387 | 0.6016 |
73
+ | 0.2187 | 21.43 | 12000 | 0.3657 | 0.5845 |
74
+ | 0.2317 | 22.32 | 12500 | 0.2932 | 0.5845 |
75
+ | 0.2091 | 23.21 | 13000 | 0.2551 | 0.5614 |
76
+ | 0.199 | 24.11 | 13500 | 0.3113 | 0.5474 |
77
+ | 0.1777 | 25.0 | 14000 | 0.2895 | 0.5572 |
78
+ | 0.1823 | 25.89 | 14500 | 0.3127 | 0.5456 |
79
+ | 0.179 | 26.79 | 15000 | 0.2945 | 0.5438 |
80
+ | 0.1596 | 27.68 | 15500 | 0.3052 | 0.5322 |
81
+ | 0.1671 | 28.57 | 16000 | 0.3119 | 0.5365 |
82
+ | 0.1564 | 29.46 | 16500 | 0.3203 | 0.5353 |
83
 
84
 
85
  ### Framework versions