Hugging Face's logo

Giseok
/
wav2vec2-base-STTTest

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Model card Files
xet
Metrics Community
1
Giseok commited on
Commit
9f1c72d
·
1 Parent(s): ac14fe2

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +87 -0
README.md ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: wav2vec2-base-STTTest
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # wav2vec2-base-STTTest
14
+
15
+ This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 0.2278
18
+ - Wer: 0.2379
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 0.0001
38
+ - train_batch_size: 8
39
+ - eval_batch_size: 8
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - lr_scheduler_warmup_steps: 1000
44
+ - num_epochs: 30
45
+ - mixed_precision_training: Native AMP
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
50
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
51
+ | 1.5239 | 1.0 | 500 | 0.3262 | 0.2950 |
52
+ | 0.655 | 2.01 | 1000 | 0.2197 | 0.2649 |
53
+ | 0.4153 | 3.01 | 1500 | 0.2049 | 0.2569 |
54
+ | 0.3449 | 4.02 | 2000 | 0.2035 | 0.2562 |
55
+ | 0.2926 | 5.02 | 2500 | 0.1979 | 0.2519 |
56
+ | 0.2597 | 6.02 | 3000 | 0.2197 | 0.2561 |
57
+ | 0.2337 | 7.03 | 3500 | 0.2711 | 0.2525 |
58
+ | 0.1829 | 8.03 | 4000 | 0.2875 | 0.2571 |
59
+ | 0.1608 | 9.04 | 4500 | 0.2398 | 0.2614 |
60
+ | 0.1902 | 10.04 | 5000 | 0.2146 | 0.2571 |
61
+ | 0.1367 | 11.04 | 5500 | 0.2411 | 0.2533 |
62
+ | 0.1465 | 12.05 | 6000 | 0.2193 | 0.2565 |
63
+ | 0.1244 | 13.05 | 6500 | 0.2626 | 0.2488 |
64
+ | 0.124 | 14.06 | 7000 | 0.2575 | 0.2507 |
65
+ | 0.111 | 15.06 | 7500 | 0.3106 | 0.2452 |
66
+ | 0.1036 | 16.06 | 8000 | 0.2525 | 0.2480 |
67
+ | 0.0989 | 17.07 | 8500 | 0.2693 | 0.2429 |
68
+ | 0.1122 | 18.07 | 9000 | 0.2569 | 0.2474 |
69
+ | 0.091 | 19.08 | 9500 | 0.2477 | 0.2432 |
70
+ | 0.0837 | 20.08 | 10000 | 0.2646 | 0.2432 |
71
+ | 0.097 | 21.08 | 10500 | 0.2547 | 0.2441 |
72
+ | 0.077 | 22.09 | 11000 | 0.2668 | 0.2417 |
73
+ | 0.0643 | 23.09 | 11500 | 0.2357 | 0.2425 |
74
+ | 0.069 | 24.1 | 12000 | 0.2372 | 0.2409 |
75
+ | 0.0598 | 25.1 | 12500 | 0.2338 | 0.2435 |
76
+ | 0.0615 | 26.1 | 13000 | 0.2369 | 0.2411 |
77
+ | 0.0762 | 27.11 | 13500 | 0.2294 | 0.2415 |
78
+ | 0.0598 | 28.11 | 14000 | 0.2284 | 0.2393 |
79
+ | 0.055 | 29.12 | 14500 | 0.2278 | 0.2379 |
80
+
81
+
82
+ ### Framework versions
83
+
84
+ - Transformers 4.17.0
85
+ - Pytorch 1.10.1+cu111
86
+ - Datasets 1.18.3
87
+ - Tokenizers 0.12.1