File size: 2,607 Bytes
5a73751
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
---
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: w2v2-base-pretrained_lr1e-4_at0.8_da0.2
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# w2v2-base-pretrained_lr1e-4_at0.8_da0.2

This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.3297
- Wer: 0.2683

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 200
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 23.6917       | 10.0  | 100  | 3.6026          | 1.0    |
| 3.259         | 20.0  | 200  | 3.1696          | 1.0    |
| 3.1152        | 30.0  | 300  | 3.1344          | 1.0    |
| 3.0799        | 40.0  | 400  | 3.0976          | 1.0    |
| 3.0793        | 50.0  | 500  | 3.0977          | 1.0    |
| 3.0692        | 60.0  | 600  | 3.0992          | 1.0    |
| 3.0604        | 70.0  | 700  | 3.1350          | 1.0    |
| 3.0397        | 80.0  | 800  | 3.0537          | 1.0    |
| 2.9847        | 90.0  | 900  | 2.9905          | 1.0    |
| 2.5926        | 100.0 | 1000 | 2.2350          | 1.0077 |
| 0.9017        | 110.0 | 1100 | 1.2152          | 0.6301 |
| 0.2326        | 120.0 | 1200 | 1.2279          | 0.4524 |
| 0.1364        | 130.0 | 1300 | 1.2103          | 0.4238 |
| 0.0935        | 140.0 | 1400 | 1.1953          | 0.3926 |
| 0.0759        | 150.0 | 1500 | 1.3237          | 0.3516 |
| 0.0599        | 160.0 | 1600 | 1.3929          | 0.3050 |
| 0.0438        | 170.0 | 1700 | 1.3132          | 0.2717 |
| 0.0389        | 180.0 | 1800 | 1.3469          | 0.2666 |
| 0.0355        | 190.0 | 1900 | 1.3029          | 0.2691 |
| 0.03          | 200.0 | 2000 | 1.3297          | 0.2683 |


### Framework versions

- Transformers 4.35.0
- Pytorch 2.0.0
- Datasets 2.14.6
- Tokenizers 0.14.1