Hugging Face's logo

DrishtiSharma
/
wav2vec2-large-xls-r-300m-or-dx12

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Oriya
wav2vec2
Generated from Trainer
hf-asr-leaderboard
model_for_talk
mozilla-foundation/common_voice_8_0
robust-speech-event
Eval Results
Inference Endpoints
Model card Files Metrics Community
DrishtiSharma commited on
Commit
2bfa38e
•
1 Parent(s): 8fac508

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +110 -0
README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - common_voice
7
+ model-index:
8
+ - name: wav2vec2-large-xls-r-300m-or-dx12
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-xls-r-300m-or-dx12
16
+
17
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 1.4638
20
+ - Wer: 0.5602
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: 0.0004
40
+ - train_batch_size: 16
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 32
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 1000
48
+ - num_epochs: 200
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:------:|:----:|:---------------:|:------:|
55
+ | 13.5059 | 4.17 | 100 | 10.3789 | 1.0 |
56
+ | 4.5964 | 8.33 | 200 | 4.3294 | 1.0 |
57
+ | 3.4448 | 12.5 | 300 | 3.7903 | 1.0 |
58
+ | 3.3683 | 16.67 | 400 | 3.5289 | 1.0 |
59
+ | 2.042 | 20.83 | 500 | 1.1531 | 0.7857 |
60
+ | 0.5721 | 25.0 | 600 | 1.0267 | 0.7646 |
61
+ | 0.3274 | 29.17 | 700 | 1.0773 | 0.6938 |
62
+ | 0.2466 | 33.33 | 800 | 1.0323 | 0.6647 |
63
+ | 0.2047 | 37.5 | 900 | 1.1255 | 0.6733 |
64
+ | 0.1847 | 41.67 | 1000 | 1.1194 | 0.6515 |
65
+ | 0.1453 | 45.83 | 1100 | 1.1215 | 0.6601 |
66
+ | 0.1367 | 50.0 | 1200 | 1.1898 | 0.6627 |
67
+ | 0.1334 | 54.17 | 1300 | 1.3082 | 0.6687 |
68
+ | 0.1041 | 58.33 | 1400 | 1.2514 | 0.6177 |
69
+ | 0.1024 | 62.5 | 1500 | 1.2055 | 0.6528 |
70
+ | 0.0919 | 66.67 | 1600 | 1.4125 | 0.6369 |
71
+ | 0.074 | 70.83 | 1700 | 1.4006 | 0.6634 |
72
+ | 0.0681 | 75.0 | 1800 | 1.3943 | 0.6131 |
73
+ | 0.0709 | 79.17 | 1900 | 1.3545 | 0.6296 |
74
+ | 0.064 | 83.33 | 2000 | 1.2437 | 0.6237 |
75
+ | 0.0552 | 87.5 | 2100 | 1.3762 | 0.6190 |
76
+ | 0.056 | 91.67 | 2200 | 1.3763 | 0.6323 |
77
+ | 0.0514 | 95.83 | 2300 | 1.2897 | 0.6164 |
78
+ | 0.0409 | 100.0 | 2400 | 1.4257 | 0.6104 |
79
+ | 0.0379 | 104.17 | 2500 | 1.4219 | 0.5853 |
80
+ | 0.0367 | 108.33 | 2600 | 1.4361 | 0.6032 |
81
+ | 0.0412 | 112.5 | 2700 | 1.4713 | 0.6098 |
82
+ | 0.0353 | 116.67 | 2800 | 1.4132 | 0.6369 |
83
+ | 0.0336 | 120.83 | 2900 | 1.5210 | 0.6098 |
84
+ | 0.0302 | 125.0 | 3000 | 1.4686 | 0.5939 |
85
+ | 0.0398 | 129.17 | 3100 | 1.5456 | 0.6204 |
86
+ | 0.0291 | 133.33 | 3200 | 1.4111 | 0.5827 |
87
+ | 0.0247 | 137.5 | 3300 | 1.3866 | 0.6151 |
88
+ | 0.0196 | 141.67 | 3400 | 1.4513 | 0.5880 |
89
+ | 0.0218 | 145.83 | 3500 | 1.5100 | 0.5899 |
90
+ | 0.0196 | 150.0 | 3600 | 1.4936 | 0.5999 |
91
+ | 0.0164 | 154.17 | 3700 | 1.5012 | 0.5701 |
92
+ | 0.0168 | 158.33 | 3800 | 1.5601 | 0.5919 |
93
+ | 0.0151 | 162.5 | 3900 | 1.4891 | 0.5761 |
94
+ | 0.0137 | 166.67 | 4000 | 1.4839 | 0.5800 |
95
+ | 0.0143 | 170.83 | 4100 | 1.4826 | 0.5754 |
96
+ | 0.0114 | 175.0 | 4200 | 1.4950 | 0.5708 |
97
+ | 0.0092 | 179.17 | 4300 | 1.5008 | 0.5694 |
98
+ | 0.0104 | 183.33 | 4400 | 1.4774 | 0.5728 |
99
+ | 0.0096 | 187.5 | 4500 | 1.4948 | 0.5767 |
100
+ | 0.0105 | 191.67 | 4600 | 1.4557 | 0.5694 |
101
+ | 0.009 | 195.83 | 4700 | 1.4615 | 0.5628 |
102
+ | 0.0081 | 200.0 | 4800 | 1.4638 | 0.5602 |
103
+
104
+
105
+ ### Framework versions
106
+
107
+ - Transformers 4.16.2
108
+ - Pytorch 1.10.0+cu111
109
+ - Datasets 1.18.3
110
+ - Tokenizers 0.11.0