jason1i commited on
Commit
485d8c1
1 Parent(s): b947f6e

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +79 -0
README.md ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - zu-HK
4
+ license: apache-2.0
5
+ tags:
6
+ - generated_from_trainer
7
+ datasets:
8
+ - mozilla-foundation/common_voice_13_0
9
+ metrics:
10
+ - wer
11
+ model-index:
12
+ - name: Whisper Small zu-HK
13
+ results:
14
+ - task:
15
+ name: Automatic Speech Recognition
16
+ type: automatic-speech-recognition
17
+ dataset:
18
+ name: Common Voice 13
19
+ type: mozilla-foundation/common_voice_13_0
20
+ config: zh-HK
21
+ split: test
22
+ args: zh-HK
23
+ metrics:
24
+ - name: Wer
25
+ type: wer
26
+ value: 64.88393977415308
27
+ ---
28
+
29
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
30
+ should probably proofread and complete it, then remove this comment. -->
31
+
32
+ # Whisper Small zu-HK
33
+
34
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 13 dataset.
35
+ It achieves the following results on the evaluation set:
36
+ - Loss: 0.2883
37
+ - Wer Ortho: 66.1207
38
+ - Wer: 64.8839
39
+
40
+ ## Model description
41
+
42
+ More information needed
43
+
44
+ ## Intended uses & limitations
45
+
46
+ More information needed
47
+
48
+ ## Training and evaluation data
49
+
50
+ More information needed
51
+
52
+ ## Training procedure
53
+
54
+ ### Training hyperparameters
55
+
56
+ The following hyperparameters were used during training:
57
+ - learning_rate: 1e-05
58
+ - train_batch_size: 16
59
+ - eval_batch_size: 16
60
+ - seed: 42
61
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
+ - lr_scheduler_type: constant_with_warmup
63
+ - lr_scheduler_warmup_steps: 50
64
+ - training_steps: 500
65
+ - mixed_precision_training: Native AMP
66
+
67
+ ### Training results
68
+
69
+ | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
70
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
71
+ | 0.3393 | 0.57 | 500 | 0.2883 | 66.1207 | 64.8839 |
72
+
73
+
74
+ ### Framework versions
75
+
76
+ - Transformers 4.30.0.dev0
77
+ - Pytorch 2.0.1+cu117
78
+ - Datasets 2.12.0
79
+ - Tokenizers 0.13.3