Hugging Face's logo

geninhu
/
xls-asr-vi-40h

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Vietnamese
wav2vec2
common-voice
hf-asr-leaderboard
robust-speech-event
Eval Results
Inference Endpoints
Model card Files Metrics Community
geninhu commited on
Commit
39461c3
·
1 Parent(s): f00e0c6

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +114 -0
README.md ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - automatic-speech-recognition
5
+ - geninhu/fpt-vi
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: xls-asr-vi-40h
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
+ # xls-asr-vi-40h
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 GENINHU/FPT-VI - NA dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 1.1177
20
+ - Wer: 0.6058
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: 5e-06
40
+ - train_batch_size: 16
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: linear
45
+ - lr_scheduler_warmup_steps: 1500
46
+ - num_epochs: 50.0
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
52
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
53
+ | 23.3878 | 0.93 | 1500 | 21.9179 | 1.0 |
54
+ | 8.8862 | 1.85 | 3000 | 6.0599 | 1.0 |
55
+ | 4.3701 | 2.78 | 4500 | 4.3837 | 1.0 |
56
+ | 4.113 | 3.7 | 6000 | 4.2698 | 0.9982 |
57
+ | 3.9666 | 4.63 | 7500 | 3.9726 | 0.9989 |
58
+ | 3.5965 | 5.56 | 9000 | 3.7124 | 0.9975 |
59
+ | 3.3944 | 6.48 | 10500 | 3.5005 | 1.0057 |
60
+ | 3.304 | 7.41 | 12000 | 3.3710 | 1.0043 |
61
+ | 3.2482 | 8.33 | 13500 | 3.4201 | 1.0155 |
62
+ | 3.212 | 9.26 | 15000 | 3.3732 | 1.0151 |
63
+ | 3.1778 | 10.19 | 16500 | 3.2763 | 1.0009 |
64
+ | 3.1027 | 11.11 | 18000 | 3.1943 | 1.0025 |
65
+ | 2.9905 | 12.04 | 19500 | 2.8082 | 0.9703 |
66
+ | 2.7095 | 12.96 | 21000 | 2.4993 | 0.9302 |
67
+ | 2.4862 | 13.89 | 22500 | 2.3072 | 0.9140 |
68
+ | 2.3271 | 14.81 | 24000 | 2.1398 | 0.8949 |
69
+ | 2.1968 | 15.74 | 25500 | 2.0594 | 0.8817 |
70
+ | 2.111 | 16.67 | 27000 | 1.9404 | 0.8630 |
71
+ | 2.0387 | 17.59 | 28500 | 1.8895 | 0.8497 |
72
+ | 1.9504 | 18.52 | 30000 | 1.7961 | 0.8315 |
73
+ | 1.9039 | 19.44 | 31500 | 1.7433 | 0.8213 |
74
+ | 1.8342 | 20.37 | 33000 | 1.6790 | 0.7994 |
75
+ | 1.7824 | 21.3 | 34500 | 1.6291 | 0.7825 |
76
+ | 1.7359 | 22.22 | 36000 | 1.5783 | 0.7706 |
77
+ | 1.7053 | 23.15 | 37500 | 1.5248 | 0.7492 |
78
+ | 1.6504 | 24.07 | 39000 | 1.4930 | 0.7406 |
79
+ | 1.6263 | 25.0 | 40500 | 1.4572 | 0.7348 |
80
+ | 1.5893 | 25.93 | 42000 | 1.4202 | 0.7161 |
81
+ | 1.5669 | 26.85 | 43500 | 1.3987 | 0.7143 |
82
+ | 1.5277 | 27.78 | 45000 | 1.3512 | 0.6991 |
83
+ | 1.501 | 28.7 | 46500 | 1.3320 | 0.6879 |
84
+ | 1.4781 | 29.63 | 48000 | 1.3112 | 0.6788 |
85
+ | 1.4477 | 30.56 | 49500 | 1.2850 | 0.6657 |
86
+ | 1.4483 | 31.48 | 51000 | 1.2813 | 0.6633 |
87
+ | 1.4065 | 32.41 | 52500 | 1.2475 | 0.6541 |
88
+ | 1.3779 | 33.33 | 54000 | 1.2244 | 0.6503 |
89
+ | 1.3788 | 34.26 | 55500 | 1.2116 | 0.6407 |
90
+ | 1.3428 | 35.19 | 57000 | 1.1938 | 0.6352 |
91
+ | 1.3453 | 36.11 | 58500 | 1.1927 | 0.6340 |
92
+ | 1.3137 | 37.04 | 60000 | 1.1699 | 0.6252 |
93
+ | 1.2984 | 37.96 | 61500 | 1.1666 | 0.6229 |
94
+ | 1.2927 | 38.89 | 63000 | 1.1585 | 0.6188 |
95
+ | 1.2919 | 39.81 | 64500 | 1.1618 | 0.6190 |
96
+ | 1.293 | 40.74 | 66000 | 1.1479 | 0.6181 |
97
+ | 1.2853 | 41.67 | 67500 | 1.1423 | 0.6202 |
98
+ | 1.2687 | 42.59 | 69000 | 1.1315 | 0.6131 |
99
+ | 1.2603 | 43.52 | 70500 | 1.1333 | 0.6128 |
100
+ | 1.2577 | 44.44 | 72000 | 1.1191 | 0.6079 |
101
+ | 1.2435 | 45.37 | 73500 | 1.1177 | 0.6079 |
102
+ | 1.251 | 46.3 | 75000 | 1.1211 | 0.6092 |
103
+ | 1.2482 | 47.22 | 76500 | 1.1177 | 0.6060 |
104
+ | 1.2422 | 48.15 | 78000 | 1.1227 | 0.6097 |
105
+ | 1.2485 | 49.07 | 79500 | 1.1187 | 0.6071 |
106
+ | 1.2425 | 50.0 | 81000 | 1.1177 | 0.6058 |
107
+
108
+
109
+ ### Framework versions
110
+
111
+ - Transformers 4.16.0.dev0
112
+ - Pytorch 1.10.1+cu102
113
+ - Datasets 1.17.1.dev0
114
+ - Tokenizers 0.11.0