Hugging Face's logo

christw16
/
wavlm-base-plus_zh_tw_ver2

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wavlm
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
christw16 commited on
Commit
f2db82d
•
1 Parent(s): 5e91740

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
+ tags:
3
+ - generated_from_trainer
4
+ datasets:
5
+ - common_voice
6
+ metrics:
7
+ - wer
8
+ model-index:
9
+ - name: wavlm-base-plus_zh_tw_ver2
10
+ results:
11
+ - task:
12
+ name: Automatic Speech Recognition
13
+ type: automatic-speech-recognition
14
+ dataset:
15
+ name: common_voice
16
+ type: common_voice
17
+ config: zh-TW
18
+ split: test
19
+ args: zh-TW
20
+ metrics:
21
+ - name: Wer
22
+ type: wer
23
+ value: 1.0
24
+ ---
25
+
26
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
27
+ should probably proofread and complete it, then remove this comment. -->
28
+
29
+ # wavlm-base-plus_zh_tw_ver2
30
+
31
+ This model is a fine-tuned version of [microsoft/wavlm-base-plus](https://huggingface.co/microsoft/wavlm-base-plus) on the common_voice dataset.
32
+ It achieves the following results on the evaluation set:
33
+ - Loss: 6.5278
34
+ - Wer: 1.0
35
+
36
+ ## Model description
37
+
38
+ More information needed
39
+
40
+ ## Intended uses & limitations
41
+
42
+ More information needed
43
+
44
+ ## Training and evaluation data
45
+
46
+ More information needed
47
+
48
+ ## Training procedure
49
+
50
+ ### Training hyperparameters
51
+
52
+ The following hyperparameters were used during training:
53
+ - learning_rate: 7.5e-05
54
+ - train_batch_size: 32
55
+ - eval_batch_size: 2
56
+ - seed: 42
57
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
58
+ - lr_scheduler_type: linear
59
+ - lr_scheduler_warmup_steps: 2000
60
+ - num_epochs: 100.0
61
+ - mixed_precision_training: Native AMP
62
+
63
+ ### Training results
64
+
65
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
66
+ |:-------------:|:-----:|:-----:|:---------------:|:---:|
67
+ | 82.628 | 2.5 | 500 | 79.5587 | 1.0 |
68
+ | 17.5635 | 5.0 | 1000 | 11.5929 | 1.0 |
69
+ | 6.4288 | 7.5 | 1500 | 6.4475 | 1.0 |
70
+ | 6.4092 | 10.0 | 2000 | 6.4579 | 1.0 |
71
+ | 6.3982 | 12.5 | 2500 | 6.4662 | 1.0 |
72
+ | 6.391 | 15.0 | 3000 | 6.4655 | 1.0 |
73
+ | 6.4097 | 17.5 | 3500 | 6.4691 | 1.0 |
74
+ | 6.3986 | 20.0 | 4000 | 6.4702 | 1.0 |
75
+ | 6.4069 | 22.5 | 4500 | 6.4761 | 1.0 |
76
+ | 6.4158 | 25.0 | 5000 | 6.4750 | 1.0 |
77
+ | 6.4117 | 27.5 | 5500 | 6.4816 | 1.0 |
78
+ | 6.4086 | 30.0 | 6000 | 6.4806 | 1.0 |
79
+ | 6.3992 | 32.5 | 6500 | 6.4872 | 1.0 |
80
+ | 6.3946 | 35.0 | 7000 | 6.4866 | 1.0 |
81
+ | 6.4212 | 37.5 | 7500 | 6.4895 | 1.0 |
82
+ | 6.4051 | 40.0 | 8000 | 6.4926 | 1.0 |
83
+ | 6.398 | 42.5 | 8500 | 6.5015 | 1.0 |
84
+ | 6.3967 | 45.0 | 9000 | 6.4960 | 1.0 |
85
+ | 6.4096 | 47.5 | 9500 | 6.5003 | 1.0 |
86
+ | 6.4068 | 50.0 | 10000 | 6.5026 | 1.0 |
87
+ | 6.4062 | 52.5 | 10500 | 6.5071 | 1.0 |
88
+ | 6.395 | 55.0 | 11000 | 6.5066 | 1.0 |
89
+ | 6.4079 | 57.5 | 11500 | 6.5093 | 1.0 |
90
+ | 6.411 | 60.0 | 12000 | 6.5106 | 1.0 |
91
+ | 6.4023 | 62.5 | 12500 | 6.5112 | 1.0 |
92
+ | 6.4053 | 65.0 | 13000 | 6.5143 | 1.0 |
93
+ | 6.4103 | 67.5 | 13500 | 6.5172 | 1.0 |
94
+ | 6.3899 | 70.0 | 14000 | 6.5182 | 1.0 |
95
+ | 6.4054 | 72.5 | 14500 | 6.5197 | 1.0 |
96
+ | 6.391 | 75.0 | 15000 | 6.5200 | 1.0 |
97
+ | 6.3988 | 77.5 | 15500 | 6.5220 | 1.0 |
98
+ | 6.4059 | 80.0 | 16000 | 6.5228 | 1.0 |
99
+ | 6.392 | 82.5 | 16500 | 6.5233 | 1.0 |
100
+ | 6.3947 | 85.0 | 17000 | 6.5253 | 1.0 |
101
+ | 6.3966 | 87.5 | 17500 | 6.5259 | 1.0 |
102
+ | 6.3905 | 90.0 | 18000 | 6.5264 | 1.0 |
103
+ | 6.4003 | 92.5 | 18500 | 6.5272 | 1.0 |
104
+ | 6.3877 | 95.0 | 19000 | 6.5275 | 1.0 |
105
+ | 6.3903 | 97.5 | 19500 | 6.5277 | 1.0 |
106
+ | 6.3944 | 100.0 | 20000 | 6.5278 | 1.0 |
107
+
108
+
109
+ ### Framework versions
110
+
111
+ - Transformers 4.28.0.dev0
112
+ - Pytorch 1.12.0+cu102
113
+ - Datasets 2.10.1
114
+ - Tokenizers 0.13.2