Hugging Face's logo

patrickvonplaten
/
wav2vec2-random

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
timit_asr
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
1
patrickvonplaten commited on
Commit
9097b44
•
1 Parent(s): ce69a58

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +90 -0
README.md ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - automatic-speech-recognition
4
+ - timit_asr
5
+ - generated_from_trainer
6
+ datasets:
7
+ - timit_asr
8
+ model-index:
9
+ - name: wav2vec2-random
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # wav2vec2-random
17
+
18
+ This model is a fine-tuned version of [patrickvonplaten/wav2vec2-base-random](https://huggingface.co/patrickvonplaten/wav2vec2-base-random) on the TIMIT_ASR - NA dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 3.1593
21
+ - Wer: 0.8364
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0001
41
+ - train_batch_size: 32
42
+ - eval_batch_size: 1
43
+ - seed: 42
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 1000
47
+ - num_epochs: 20.0
48
+ - mixed_precision_training: Native AMP
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
53
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
54
+ | 2.9043 | 0.69 | 100 | 2.9683 | 1.0 |
55
+ | 2.8537 | 1.38 | 200 | 2.9281 | 0.9997 |
56
+ | 2.7803 | 2.07 | 300 | 2.7330 | 0.9999 |
57
+ | 2.6806 | 2.76 | 400 | 2.5792 | 1.0 |
58
+ | 2.4136 | 3.45 | 500 | 2.4327 | 0.9948 |
59
+ | 2.1682 | 4.14 | 600 | 2.3508 | 0.9877 |
60
+ | 2.2577 | 4.83 | 700 | 2.2176 | 0.9773 |
61
+ | 2.355 | 5.52 | 800 | 2.1753 | 0.9542 |
62
+ | 1.8588 | 6.21 | 900 | 2.0650 | 0.8851 |
63
+ | 1.6831 | 6.9 | 1000 | 2.0109 | 0.8618 |
64
+ | 1.888 | 7.59 | 1100 | 1.9660 | 0.8418 |
65
+ | 2.0066 | 8.28 | 1200 | 1.9847 | 0.8531 |
66
+ | 1.7044 | 8.97 | 1300 | 1.9760 | 0.8527 |
67
+ | 1.3168 | 9.66 | 1400 | 2.0708 | 0.8327 |
68
+ | 1.2143 | 10.34 | 1500 | 2.0601 | 0.8419 |
69
+ | 1.6189 | 11.03 | 1600 | 2.0960 | 0.8299 |
70
+ | 1.13 | 11.72 | 1700 | 2.2540 | 0.8408 |
71
+ | 0.8001 | 12.41 | 1800 | 2.4260 | 0.8306 |
72
+ | 0.7769 | 13.1 | 1900 | 2.4182 | 0.8445 |
73
+ | 1.2165 | 13.79 | 2000 | 2.3666 | 0.8284 |
74
+ | 0.8026 | 14.48 | 2100 | 2.7118 | 0.8662 |
75
+ | 0.5148 | 15.17 | 2200 | 2.7957 | 0.8526 |
76
+ | 0.4921 | 15.86 | 2300 | 2.8244 | 0.8346 |
77
+ | 0.7629 | 16.55 | 2400 | 2.8944 | 0.8370 |
78
+ | 0.5762 | 17.24 | 2500 | 3.0335 | 0.8367 |
79
+ | 0.4076 | 17.93 | 2600 | 3.0776 | 0.8358 |
80
+ | 0.3395 | 18.62 | 2700 | 3.1572 | 0.8261 |
81
+ | 0.4862 | 19.31 | 2800 | 3.1319 | 0.8414 |
82
+ | 0.5061 | 20.0 | 2900 | 3.1593 | 0.8364 |
83
+
84
+
85
+ ### Framework versions
86
+
87
+ - Transformers 4.12.0.dev0
88
+ - Pytorch 1.8.1
89
+ - Datasets 1.14.1.dev0
90
+ - Tokenizers 0.10.3