File size: 2,308 Bytes
7773fb6
ffee101
7773fb6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ffee101
7773fb6
 
 
 
 
 
 
 
 
ffee101
 
7773fb6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ffee101
7773fb6
 
 
 
 
 
 
ffee101
 
7773fb6
 
 
 
 
 
ffee101
 
 
 
 
 
7773fb6
 
 
 
ffee101
7773fb6
2d82c21
7773fb6
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
83
84
85
86
87
88
---
library_name: transformers
language:
- gn
license: apache-2.0
base_model: glob-asr/wav2vec2-large-xls-r-300m-guarani-small
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_1
metrics:
- wer
model-index:
- name: Common Voice 16
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: Common Voice 16
      type: mozilla-foundation/common_voice_16_1
      config: gn
      split: None
      args: gn
    metrics:
    - name: Wer
      type: wer
      value: 39.84010659560293
---

<!-- 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. -->

# Common Voice 16

This model is a fine-tuned version of [glob-asr/wav2vec2-large-xls-r-300m-guarani-small](https://huggingface.co/glob-asr/wav2vec2-large-xls-r-300m-guarani-small) on the Common Voice 16 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2438
- Wer: 39.8401

## 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: 8
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 3000
- training_steps: 3000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer     |
|:-------------:|:------:|:----:|:---------------:|:-------:|
| 1.2579        | 0.4955 | 500  | 0.3710          | 53.4310 |
| 0.919         | 0.9911 | 1000 | 0.3295          | 49.9001 |
| 0.746         | 1.4866 | 1500 | 0.2902          | 45.1033 |
| 0.6767        | 1.9822 | 2000 | 0.2674          | 43.3711 |
| 0.574         | 2.4777 | 2500 | 0.2677          | 42.5716 |
| 0.5485        | 2.9732 | 3000 | 0.2438          | 39.8401 |


### Framework versions

- Transformers 4.44.1
- Pytorch 2.3.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1