Hugging Face's logo

rafiulrumy
/
wav2vec2-large-xlsr-hindi-demo-colab_2

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
generated_from_trainer
Inference Endpoints
Model card Files Metrics Community
rafiulrumy commited on
Commit
67c4023
•
1 Parent(s): ab35704

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +107 -0
README.md ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - common_voice
7
+ model-index:
8
+ - name: wav2vec2-large-xlsr-hindi-demo-colab_2
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # wav2vec2-large-xlsr-hindi-demo-colab_2
16
+
17
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the common_voice dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 3.8793
20
+ - Wer: 1.1357
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 0.0003
40
+ - train_batch_size: 8
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 16
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 20
48
+ - num_epochs: 50
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
55
+ | 22.381 | 1.11 | 20 | 22.1964 | 1.0 |
56
+ | 7.6212 | 2.22 | 40 | 4.0591 | 1.0 |
57
+ | 3.6951 | 3.32 | 60 | 3.6782 | 1.0 |
58
+ | 3.5574 | 4.43 | 80 | 3.6776 | 1.0 |
59
+ | 3.5374 | 5.54 | 100 | 3.5649 | 1.0 |
60
+ | 3.5512 | 6.65 | 120 | 3.5266 | 1.0 |
61
+ | 3.5075 | 7.76 | 140 | 3.6860 | 1.0 |
62
+ | 3.5097 | 8.86 | 160 | 3.4941 | 1.0 |
63
+ | 3.481 | 9.97 | 180 | 3.4659 | 1.0 |
64
+ | 3.5623 | 11.11 | 200 | 3.7254 | 1.0 |
65
+ | 3.4404 | 12.22 | 220 | 3.5225 | 1.0 |
66
+ | 3.432 | 13.32 | 240 | 3.5706 | 1.0 |
67
+ | 3.4177 | 14.43 | 260 | 3.3833 | 1.0 |
68
+ | 3.3735 | 15.54 | 280 | 3.4140 | 1.0 |
69
+ | 3.31 | 16.65 | 300 | 3.2702 | 1.0 |
70
+ | 3.2256 | 17.76 | 320 | 3.2405 | 1.0 |
71
+ | 3.0546 | 18.86 | 340 | 3.1644 | 1.0 |
72
+ | 2.7233 | 19.97 | 360 | 2.9753 | 1.0 |
73
+ | 2.2822 | 21.11 | 380 | 3.1119 | 1.1183 |
74
+ | 1.8027 | 22.22 | 400 | 3.0035 | 1.2378 |
75
+ | 1.5274 | 23.32 | 420 | 2.8536 | 1.2227 |
76
+ | 1.2313 | 24.43 | 440 | 2.9544 | 1.0951 |
77
+ | 1.0956 | 25.54 | 460 | 2.8814 | 1.0661 |
78
+ | 0.9456 | 26.65 | 480 | 3.1192 | 1.1589 |
79
+ | 0.7893 | 27.76 | 500 | 3.2919 | 1.1833 |
80
+ | 0.7256 | 28.86 | 520 | 3.0864 | 1.0951 |
81
+ | 0.6051 | 29.97 | 540 | 3.5888 | 1.1821 |
82
+ | 0.6087 | 31.11 | 560 | 3.4579 | 1.1392 |
83
+ | 0.5529 | 32.22 | 580 | 3.1998 | 1.0708 |
84
+ | 0.5211 | 33.32 | 600 | 3.4655 | 1.1311 |
85
+ | 0.4506 | 34.43 | 620 | 3.4338 | 1.1694 |
86
+ | 0.4101 | 35.54 | 640 | 3.5189 | 1.1450 |
87
+ | 0.4484 | 36.65 | 660 | 3.6585 | 1.1601 |
88
+ | 0.4038 | 37.76 | 680 | 3.6314 | 1.1497 |
89
+ | 0.3539 | 38.86 | 700 | 3.6955 | 1.1485 |
90
+ | 0.3898 | 39.97 | 720 | 3.5738 | 1.1148 |
91
+ | 0.35 | 41.11 | 740 | 3.6594 | 1.1195 |
92
+ | 0.3328 | 42.22 | 760 | 3.6894 | 1.1299 |
93
+ | 0.3264 | 43.32 | 780 | 3.7290 | 1.1021 |
94
+ | 0.3364 | 44.43 | 800 | 3.7256 | 1.1543 |
95
+ | 0.3071 | 45.54 | 820 | 3.8834 | 1.1415 |
96
+ | 0.3074 | 46.65 | 840 | 3.8077 | 1.1450 |
97
+ | 0.3064 | 47.76 | 860 | 3.8733 | 1.1346 |
98
+ | 0.3223 | 48.86 | 880 | 3.8780 | 1.1323 |
99
+ | 0.275 | 49.97 | 900 | 3.8793 | 1.1357 |
100
+
101
+
102
+ ### Framework versions
103
+
104
+ - Transformers 4.11.3
105
+ - Pytorch 1.10.0+cu111
106
+ - Datasets 1.13.3
107
+ - Tokenizers 0.10.3