Hugging Face's logo

patrickvonplaten
/
wav2vec2-base-timit-fine-tuned

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
timit_asr
generated_from_trainer
Inference Endpoints
Model card Files Metrics Community
3
patrickvonplaten commited on
Commit
1f4008e
•
1 Parent(s): 8d7a280

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +91 -0
README.md ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - automatic-speech-recognition
5
+ - timit_asr
6
+ - generated_from_trainer
7
+ datasets:
8
+ - timit_asr
9
+ model-index:
10
+ - name: wav2vec2-base-timit-fine-tuned
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # wav2vec2-base-timit-fine-tuned
18
+
19
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the TIMIT_ASR - NA dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.3457
22
+ - Wer: 0.2151
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 0.0001
42
+ - train_batch_size: 32
43
+ - eval_batch_size: 1
44
+ - seed: 42
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 1000
48
+ - num_epochs: 20.0
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
55
+ | 3.1621 | 0.69 | 100 | 3.1102 | 1.0 |
56
+ | 2.9592 | 1.38 | 200 | 2.9603 | 1.0 |
57
+ | 2.9116 | 2.07 | 300 | 2.8921 | 1.0 |
58
+ | 2.1332 | 2.76 | 400 | 1.9718 | 0.9958 |
59
+ | 0.8477 | 3.45 | 500 | 0.7813 | 0.5237 |
60
+ | 0.4251 | 4.14 | 600 | 0.5166 | 0.3982 |
61
+ | 0.3743 | 4.83 | 700 | 0.4400 | 0.3578 |
62
+ | 0.4194 | 5.52 | 800 | 0.4077 | 0.3370 |
63
+ | 0.23 | 6.21 | 900 | 0.4018 | 0.3142 |
64
+ | 0.1554 | 6.9 | 1000 | 0.3623 | 0.2995 |
65
+ | 0.1511 | 7.59 | 1100 | 0.3433 | 0.2697 |
66
+ | 0.1983 | 8.28 | 1200 | 0.3539 | 0.2715 |
67
+ | 0.1443 | 8.97 | 1300 | 0.3622 | 0.2551 |
68
+ | 0.0971 | 9.66 | 1400 | 0.3580 | 0.2519 |
69
+ | 0.0764 | 10.34 | 1500 | 0.3529 | 0.2437 |
70
+ | 0.1203 | 11.03 | 1600 | 0.3455 | 0.2431 |
71
+ | 0.0881 | 11.72 | 1700 | 0.3648 | 0.2415 |
72
+ | 0.0521 | 12.41 | 1800 | 0.3564 | 0.2320 |
73
+ | 0.0434 | 13.1 | 1900 | 0.3485 | 0.2270 |
74
+ | 0.0864 | 13.79 | 2000 | 0.3517 | 0.2228 |
75
+ | 0.0651 | 14.48 | 2100 | 0.3506 | 0.2285 |
76
+ | 0.0423 | 15.17 | 2200 | 0.3428 | 0.2247 |
77
+ | 0.0302 | 15.86 | 2300 | 0.3372 | 0.2198 |
78
+ | 0.0548 | 16.55 | 2400 | 0.3496 | 0.2196 |
79
+ | 0.0674 | 17.24 | 2500 | 0.3407 | 0.2166 |
80
+ | 0.0291 | 17.93 | 2600 | 0.3512 | 0.2171 |
81
+ | 0.0298 | 18.62 | 2700 | 0.3363 | 0.2158 |
82
+ | 0.0419 | 19.31 | 2800 | 0.3493 | 0.2145 |
83
+ | 0.046 | 20.0 | 2900 | 0.3457 | 0.2151 |
84
+
85
+
86
+ ### Framework versions
87
+
88
+ - Transformers 4.12.0.dev0
89
+ - Pytorch 1.8.1
90
+ - Datasets 1.14.1.dev0
91
+ - Tokenizers 0.10.3