Hugging Face's logo

Kuray107
/
swbd-5percent-supervised

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Model card Files
xet
 Community
Kuray107 commited on
Commit
cabf9cd
·
1 Parent(s): f71e6c8

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +104 -0
README.md ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: swbd-5percent-supervised
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # swbd-5percent-supervised
14
+
15
+ This model is a fine-tuned version of [facebook/wav2vec2-large-lv60](https://huggingface.co/facebook/wav2vec2-large-lv60) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 0.6970
18
+ - Wer: 0.1352
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 0.0001
38
+ - train_batch_size: 8
39
+ - eval_batch_size: 8
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - lr_scheduler_warmup_steps: 1000
44
+ - num_epochs: 30
45
+ - mixed_precision_training: Native AMP
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
50
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
51
+ | 6.8534 | 0.64 | 1000 | 2.9535 | 1.0 |
52
+ | 1.8605 | 1.28 | 2000 | 0.7878 | 0.3719 |
53
+ | 0.9862 | 1.92 | 3000 | 0.5906 | 0.2684 |
54
+ | 0.8405 | 2.56 | 4000 | 0.5555 | 0.2151 |
55
+ | 0.6972 | 3.2 | 5000 | 0.5905 | 0.1992 |
56
+ | 0.6033 | 3.84 | 6000 | 0.4867 | 0.1781 |
57
+ | 0.5393 | 4.48 | 7000 | 0.5447 | 0.1805 |
58
+ | 0.529 | 5.12 | 8000 | 0.5398 | 0.1746 |
59
+ | 0.5072 | 5.77 | 9000 | 0.5093 | 0.1706 |
60
+ | 0.4331 | 6.41 | 10000 | 0.4990 | 0.1627 |
61
+ | 0.4837 | 7.05 | 11000 | 0.5319 | 0.1634 |
62
+ | 0.3867 | 7.69 | 12000 | 0.4866 | 0.1595 |
63
+ | 0.345 | 8.33 | 13000 | 0.5202 | 0.1582 |
64
+ | 0.372 | 8.97 | 14000 | 0.5396 | 0.1547 |
65
+ | 0.355 | 9.61 | 15000 | 0.5992 | 0.1493 |
66
+ | 0.3258 | 10.25 | 16000 | 0.5247 | 0.1527 |
67
+ | 0.3327 | 10.89 | 17000 | 0.5664 | 0.1512 |
68
+ | 0.3422 | 11.53 | 18000 | 0.5819 | 0.1456 |
69
+ | 0.2815 | 12.17 | 19000 | 0.5692 | 0.1453 |
70
+ | 0.2719 | 12.81 | 20000 | 0.5012 | 0.1476 |
71
+ | 0.2838 | 13.45 | 21000 | 0.5286 | 0.1454 |
72
+ | 0.2418 | 14.09 | 22000 | 0.6238 | 0.1486 |
73
+ | 0.2412 | 14.73 | 23000 | 0.5889 | 0.1456 |
74
+ | 0.2227 | 15.37 | 24000 | 0.5901 | 0.1459 |
75
+ | 0.2129 | 16.02 | 25000 | 0.5959 | 0.1454 |
76
+ | 0.2071 | 16.66 | 26000 | 0.6259 | 0.1427 |
77
+ | 0.2185 | 17.3 | 27000 | 0.6581 | 0.1437 |
78
+ | 0.1982 | 17.94 | 28000 | 0.6194 | 0.1411 |
79
+ | 0.1928 | 18.58 | 29000 | 0.5940 | 0.1409 |
80
+ | 0.1885 | 19.22 | 30000 | 0.6733 | 0.1417 |
81
+ | 0.1835 | 19.86 | 31000 | 0.6363 | 0.1393 |
82
+ | 0.1756 | 20.5 | 32000 | 0.6675 | 0.1382 |
83
+ | 0.1776 | 21.14 | 33000 | 0.6147 | 0.1407 |
84
+ | 0.1758 | 21.78 | 34000 | 0.6405 | 0.1420 |
85
+ | 0.1645 | 22.42 | 35000 | 0.6999 | 0.1401 |
86
+ | 0.1631 | 23.06 | 36000 | 0.6224 | 0.1385 |
87
+ | 0.1494 | 23.7 | 37000 | 0.6639 | 0.1374 |
88
+ | 0.1472 | 24.34 | 38000 | 0.6471 | 0.1373 |
89
+ | 0.1514 | 24.98 | 39000 | 0.6570 | 0.1395 |
90
+ | 0.1527 | 25.62 | 40000 | 0.6876 | 0.1375 |
91
+ | 0.1514 | 26.27 | 41000 | 0.6835 | 0.1376 |
92
+ | 0.1344 | 26.91 | 42000 | 0.6987 | 0.1372 |
93
+ | 0.1267 | 27.55 | 43000 | 0.7026 | 0.1362 |
94
+ | 0.1384 | 28.19 | 44000 | 0.7021 | 0.1366 |
95
+ | 0.1264 | 28.83 | 45000 | 0.7016 | 0.1355 |
96
+ | 0.1227 | 29.47 | 46000 | 0.6970 | 0.1352 |
97
+
98
+
99
+ ### Framework versions
100
+
101
+ - Transformers 4.14.1
102
+ - Pytorch 1.10.2
103
+ - Datasets 1.18.2
104
+ - Tokenizers 0.10.3