Hugging Face's logo

EngNada
/
wav2vec2-large-xlsr-53-demo1

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Model card Files Metrics Community
1
EngNada commited on
Commit
5ebc3af
·
1 Parent(s): 4432cf4

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-53-demo1
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-53-demo1
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: 0.9692
20
+ - Wer: 0.8462
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: 5
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 10
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 100
48
+ - num_epochs: 3
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
53
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
54
+ | 12.978 | 0.06 | 100 | 3.5377 | 1.0 |
55
+ | 3.5026 | 0.13 | 200 | 3.4366 | 1.0 |
56
+ | 3.4084 | 0.19 | 300 | 3.3831 | 1.0 |
57
+ | 3.3551 | 0.26 | 400 | 3.2563 | 1.0 |
58
+ | 3.2668 | 0.32 | 500 | 3.2109 | 1.0 |
59
+ | 2.9398 | 0.38 | 600 | 2.4548 | 0.9987 |
60
+ | 2.2204 | 0.45 | 700 | 1.8870 | 1.0135 |
61
+ | 1.7401 | 0.51 | 800 | 1.6816 | 1.0247 |
62
+ | 1.5748 | 0.57 | 900 | 1.4741 | 0.9953 |
63
+ | 1.4539 | 0.64 | 1000 | 1.4573 | 0.9852 |
64
+ | 1.3612 | 0.7 | 1100 | 1.3534 | 0.9529 |
65
+ | 1.3328 | 0.77 | 1200 | 1.3380 | 0.9320 |
66
+ | 1.2459 | 0.83 | 1300 | 1.2984 | 0.9247 |
67
+ | 1.1976 | 0.89 | 1400 | 1.2515 | 0.9252 |
68
+ | 1.1593 | 0.96 | 1500 | 1.2345 | 0.9030 |
69
+ | 1.1094 | 1.02 | 1600 | 1.2135 | 0.9305 |
70
+ | 1.0485 | 1.09 | 1700 | 1.2045 | 0.9121 |
71
+ | 0.9893 | 1.15 | 1800 | 1.1876 | 0.8990 |
72
+ | 1.0099 | 1.21 | 1900 | 1.1663 | 0.8889 |
73
+ | 0.982 | 1.28 | 2000 | 1.1674 | 0.8901 |
74
+ | 0.9975 | 1.34 | 2100 | 1.1181 | 0.8812 |
75
+ | 0.952 | 1.4 | 2200 | 1.1119 | 0.8817 |
76
+ | 0.9311 | 1.47 | 2300 | 1.0786 | 0.8773 |
77
+ | 0.9398 | 1.53 | 2400 | 1.1016 | 0.8720 |
78
+ | 0.9148 | 1.6 | 2500 | 1.0878 | 0.8778 |
79
+ | 0.9114 | 1.66 | 2600 | 1.1004 | 0.8712 |
80
+ | 0.902 | 1.72 | 2700 | 1.0223 | 0.8744 |
81
+ | 0.8978 | 1.79 | 2800 | 1.0616 | 0.8459 |
82
+ | 0.8675 | 1.85 | 2900 | 1.0974 | 0.8643 |
83
+ | 0.8373 | 1.92 | 3000 | 1.0389 | 0.8547 |
84
+ | 0.8575 | 1.98 | 3100 | 1.0388 | 0.8480 |
85
+ | 0.8313 | 2.04 | 3200 | 1.0001 | 0.8648 |
86
+ | 0.7357 | 2.11 | 3300 | 1.0222 | 0.8705 |
87
+ | 0.743 | 2.17 | 3400 | 1.0859 | 0.8765 |
88
+ | 0.7306 | 2.23 | 3500 | 1.0109 | 0.8515 |
89
+ | 0.7525 | 2.3 | 3600 | 0.9942 | 0.8619 |
90
+ | 0.7308 | 2.36 | 3700 | 1.0004 | 0.8578 |
91
+ | 0.7266 | 2.43 | 3800 | 1.0003 | 0.8497 |
92
+ | 0.737 | 2.49 | 3900 | 1.0146 | 0.8505 |
93
+ | 0.7202 | 2.55 | 4000 | 1.0172 | 0.8653 |
94
+ | 0.6945 | 2.62 | 4100 | 0.9894 | 0.8415 |
95
+ | 0.6633 | 2.68 | 4200 | 0.9894 | 0.8496 |
96
+ | 0.6972 | 2.75 | 4300 | 0.9805 | 0.8505 |
97
+ | 0.6872 | 2.81 | 4400 | 0.9939 | 0.8509 |
98
+ | 0.7238 | 2.87 | 4500 | 0.9740 | 0.8532 |
99
+ | 0.6847 | 2.94 | 4600 | 0.9692 | 0.8462 |
100
+
101
+
102
+ ### Framework versions
103
+
104
+ - Transformers 4.11.3
105
+ - Pytorch 1.10.0+cu111
106
+ - Datasets 1.14.0
107
+ - Tokenizers 0.10.3