File size: 2,110 Bytes
06e5257
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: cc-by-nc-4.0
base_model: facebook/mms-1b-all
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_13_0
metrics:
- wer
model-index:
- name: wav2vec2-large-mms-1b-por
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: mozilla-foundation/common_voice_13_0
      type: mozilla-foundation/common_voice_13_0
      config: pt
      split: test
      args: pt
    metrics:
    - name: Wer
      type: wer
      value: 0.11407164830802818
---

<!-- 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. -->

# wav2vec2-large-mms-1b-por

This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the mozilla-foundation/common_voice_13_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1340
- Wer: 0.1141

## 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: 5e-05
- 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: constant_with_warmup
- lr_scheduler_warmup_steps: 100
- training_steps: 2500
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.3219        | 0.55  | 500  | 0.1743          | 0.1302 |
| 0.2443        | 1.1   | 1000 | 0.1480          | 0.1206 |
| 0.2358        | 1.65  | 1500 | 0.1402          | 0.1167 |
| 0.223         | 2.21  | 2000 | 0.1364          | 0.1159 |
| 0.2213        | 2.76  | 2500 | 0.1340          | 0.1141 |


### Framework versions

- Transformers 4.35.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.14.1