Hugging Face's logo

bgstud
/
whisper-small-en

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
whisper
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
1
bgstud commited on
Commit
8841cba
•
1 Parent(s): 19051ac

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +97 -0
README.md ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - librispeech_asr
7
+ metrics:
8
+ - wer
9
+ model-index:
10
+ - name: whisper-small-en
11
+ results:
12
+ - task:
13
+ name: Automatic Speech Recognition
14
+ type: automatic-speech-recognition
15
+ dataset:
16
+ name: librispeech_asr
17
+ type: librispeech_asr
18
+ config: clean
19
+ split: test
20
+ args: clean
21
+ metrics:
22
+ - name: Wer
23
+ type: wer
24
+ value: 124.51154529307283
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # whisper-small-en
31
+
32
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the librispeech_asr dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 6.7832
35
+ - Wer: 124.5115
36
+
37
+ ## Model description
38
+
39
+ More information needed
40
+
41
+ ## Intended uses & limitations
42
+
43
+ More information needed
44
+
45
+ ## Training and evaluation data
46
+
47
+ More information needed
48
+
49
+ ## Training procedure
50
+
51
+ ### Training hyperparameters
52
+
53
+ The following hyperparameters were used during training:
54
+ - learning_rate: 0.0005
55
+ - train_batch_size: 16
56
+ - eval_batch_size: 8
57
+ - seed: 42
58
+ - gradient_accumulation_steps: 2
59
+ - total_train_batch_size: 32
60
+ - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
61
+ - lr_scheduler_type: linear
62
+ - lr_scheduler_warmup_steps: 2
63
+ - training_steps: 100
64
+ - mixed_precision_training: Native AMP
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
69
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|
70
+ | 9.6259 | 1.57 | 5 | 10.7408 | 1127.3535 |
71
+ | 11.5288 | 3.29 | 10 | 9.2534 | 100.0 |
72
+ | 10.9249 | 4.86 | 15 | 7.8357 | 100.0 |
73
+ | 7.0442 | 6.57 | 20 | 6.9971 | 595.3819 |
74
+ | 8.6762 | 8.29 | 25 | 5.6135 | 312.2558 |
75
+ | 5.4239 | 9.86 | 30 | 5.4885 | 97.1581 |
76
+ | 4.986 | 11.57 | 35 | 5.2888 | 628.7744 |
77
+ | 6.708 | 13.29 | 40 | 4.9665 | 277.6199 |
78
+ | 3.9096 | 14.86 | 45 | 5.0861 | 631.9716 |
79
+ | 3.2326 | 16.57 | 50 | 5.0090 | 279.7513 |
80
+ | 3.9691 | 18.29 | 55 | 5.0804 | 133.2149 |
81
+ | 1.8661 | 19.86 | 60 | 5.4423 | 317.5844 |
82
+ | 1.1588 | 21.57 | 65 | 5.7955 | 119.5382 |
83
+ | 1.0355 | 23.29 | 70 | 6.0458 | 190.2309 |
84
+ | 0.3455 | 24.86 | 75 | 6.3057 | 106.7496 |
85
+ | 0.142 | 26.57 | 80 | 6.5767 | 209.9467 |
86
+ | 0.1722 | 28.29 | 85 | 6.5937 | 101.4210 |
87
+ | 0.0816 | 29.86 | 90 | 6.7679 | 149.7336 |
88
+ | 0.079 | 31.57 | 95 | 6.8008 | 133.5702 |
89
+ | 0.1007 | 33.29 | 100 | 6.7832 | 124.5115 |
90
+
91
+
92
+ ### Framework versions
93
+
94
+ - Transformers 4.26.0.dev0
95
+ - Pytorch 1.12.1+cu113
96
+ - Datasets 2.7.1
97
+ - Tokenizers 0.13.2