Hugging Face's logo

lucio
/
xls-r-uyghur-cv7

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
Uyghur
wav2vec2
mozilla-foundation/common_voice_7_0
Generated from Trainer
robust-speech-event
hf-asr-leaderboard
Eval Results
Inference Endpoints
Model card Files Metrics Community
4
lucio commited on
Commit
27b26a4
•
1 Parent(s): 206bed3

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +44 -26
README.md CHANGED
@@ -9,19 +9,19 @@ tags:
9
  datasets:
10
  - common_voice
11
  model-index:
12
- - name: uyghur
13
  results: []
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
- # uyghur
20
 
21
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - UG dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.2266
24
- - Wer: 0.3655
25
 
26
  ## Model description
27
 
@@ -40,7 +40,7 @@ More information needed
40
  ### Training hyperparameters
41
 
42
  The following hyperparameters were used during training:
43
- - learning_rate: 7.5e-05
44
  - train_batch_size: 8
45
  - eval_batch_size: 8
46
  - seed: 42
@@ -49,31 +49,49 @@ The following hyperparameters were used during training:
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
  - lr_scheduler_warmup_steps: 2000
52
- - num_epochs: 50.0
53
  - mixed_precision_training: Native AMP
54
 
55
  ### Training results
56
 
57
- | Training Loss | Epoch | Step | Validation Loss | Wer |
58
- |:-------------:|:-----:|:----:|:---------------:|:------:|
59
- | 3.6863 | 2.73 | 500 | 3.5362 | 1.0 |
60
- | 3.1409 | 5.46 | 1000 | 3.1328 | 1.0 |
61
- | 1.8979 | 8.2 | 1500 | 0.9715 | 0.8864 |
62
- | 1.4859 | 10.93 | 2000 | 0.5234 | 0.7063 |
63
- | 1.3388 | 13.66 | 2500 | 0.4094 | 0.6203 |
64
- | 1.2531 | 16.39 | 3000 | 0.3596 | 0.5185 |
65
- | 1.1992 | 19.13 | 3500 | 0.3221 | 0.4854 |
66
- | 1.1589 | 21.86 | 4000 | 0.3040 | 0.4610 |
67
- | 1.1345 | 24.59 | 4500 | 0.2907 | 0.4450 |
68
- | 1.086 | 27.32 | 5000 | 0.2744 | 0.4299 |
69
- | 1.0697 | 30.05 | 5500 | 0.2617 | 0.4148 |
70
- | 1.0518 | 32.79 | 6000 | 0.2563 | 0.4033 |
71
- | 1.0101 | 35.52 | 6500 | 0.2480 | 0.3934 |
72
- | 1.0013 | 38.25 | 7000 | 0.2412 | 0.3855 |
73
- | 0.9845 | 40.98 | 7500 | 0.2397 | 0.3771 |
74
- | 0.9739 | 43.72 | 8000 | 0.2303 | 0.3726 |
75
- | 0.9636 | 46.45 | 8500 | 0.2285 | 0.3687 |
76
- | 0.9466 | 49.18 | 9000 | 0.2261 | 0.3648 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
77
 
78
 
79
  ### Framework versions
 
9
  datasets:
10
  - common_voice
11
  model-index:
12
+ - name: xls-r-uyghur-cv7
13
  results: []
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
+ # xls-r-uyghur-cv7
20
 
21
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - UG dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.1772
24
+ - Wer: 0.2589
25
 
26
  ## Model description
27
 
 
40
  ### Training hyperparameters
41
 
42
  The following hyperparameters were used during training:
43
+ - learning_rate: 0.0001
44
  - train_batch_size: 8
45
  - eval_batch_size: 8
46
  - seed: 42
 
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
  - lr_scheduler_warmup_steps: 2000
52
+ - num_epochs: 100.0
53
  - mixed_precision_training: Native AMP
54
 
55
  ### Training results
56
 
57
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
58
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
59
+ | 3.3043 | 2.73 | 500 | 3.2415 | 1.0 |
60
+ | 3.0482 | 5.46 | 1000 | 2.9591 | 1.0 |
61
+ | 1.4767 | 8.2 | 1500 | 0.4779 | 0.5777 |
62
+ | 1.3152 | 10.93 | 2000 | 0.3697 | 0.4938 |
63
+ | 1.2246 | 13.66 | 2500 | 0.3084 | 0.4459 |
64
+ | 1.1781 | 16.39 | 3000 | 0.2842 | 0.4154 |
65
+ | 1.1351 | 19.13 | 3500 | 0.2615 | 0.3929 |
66
+ | 1.1052 | 21.86 | 4000 | 0.2462 | 0.3747 |
67
+ | 1.0711 | 24.59 | 4500 | 0.2366 | 0.3652 |
68
+ | 1.035 | 27.32 | 5000 | 0.2268 | 0.3557 |
69
+ | 1.0277 | 30.05 | 5500 | 0.2243 | 0.3450 |
70
+ | 1.002 | 32.79 | 6000 | 0.2204 | 0.3389 |
71
+ | 0.9837 | 35.52 | 6500 | 0.2156 | 0.3349 |
72
+ | 0.9773 | 38.25 | 7000 | 0.2127 | 0.3289 |
73
+ | 0.9807 | 40.98 | 7500 | 0.2142 | 0.3274 |
74
+ | 0.9582 | 43.72 | 8000 | 0.2004 | 0.3142 |
75
+ | 0.9548 | 46.45 | 8500 | 0.2022 | 0.3050 |
76
+ | 0.9251 | 49.18 | 9000 | 0.2019 | 0.3035 |
77
+ | 0.9103 | 51.91 | 9500 | 0.1964 | 0.3021 |
78
+ | 0.915 | 54.64 | 10000 | 0.1970 | 0.3032 |
79
+ | 0.8962 | 57.38 | 10500 | 0.2007 | 0.3046 |
80
+ | 0.8729 | 60.11 | 11000 | 0.1967 | 0.2942 |
81
+ | 0.8744 | 62.84 | 11500 | 0.1952 | 0.2885 |
82
+ | 0.874 | 65.57 | 12000 | 0.1894 | 0.2895 |
83
+ | 0.8457 | 68.31 | 12500 | 0.1895 | 0.2828 |
84
+ | 0.8519 | 71.04 | 13000 | 0.1912 | 0.2875 |
85
+ | 0.8301 | 73.77 | 13500 | 0.1878 | 0.2760 |
86
+ | 0.8226 | 76.5 | 14000 | 0.1808 | 0.2701 |
87
+ | 0.8071 | 79.23 | 14500 | 0.1849 | 0.2741 |
88
+ | 0.7999 | 81.97 | 15000 | 0.1808 | 0.2717 |
89
+ | 0.7947 | 84.7 | 15500 | 0.1821 | 0.2716 |
90
+ | 0.7783 | 87.43 | 16000 | 0.1824 | 0.2661 |
91
+ | 0.7729 | 90.16 | 16500 | 0.1773 | 0.2639 |
92
+ | 0.7759 | 92.9 | 17000 | 0.1767 | 0.2629 |
93
+ | 0.7713 | 95.63 | 17500 | 0.1780 | 0.2621 |
94
+ | 0.7628 | 98.36 | 18000 | 0.1773 | 0.2594 |
95
 
96
 
97
  ### Framework versions