File size: 1,877 Bytes
65da849
f80f566
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
65da849
 
f80f566
 
65da849
f80f566
65da849
f80f566
 
 
 
65da849
f80f566
65da849
f80f566
65da849
f80f566
65da849
f80f566
65da849
f80f566
65da849
f80f566
65da849
f80f566
65da849
f80f566
65da849
f80f566
 
 
 
 
 
 
 
 
 
 
 
65da849
f80f566
65da849
f80f566
 
 
 
65da849
 
f80f566
65da849
f80f566
 
 
 
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
---
license: mit
base_model: facebook/w2v-bert-2.0
tags:
- generated_from_trainer
datasets:
- common_voice_16_0
metrics:
- wer
model-index:
- name: w2v-bert-yoruba-CV16
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: common_voice_16_0
      type: common_voice_16_0
      config: yo
      split: test
      args: yo
    metrics:
    - name: Wer
      type: wer
      value: 0.6474401754218705
---

<!-- 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. -->

# w2v-bert-yoruba-CV16

This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the common_voice_16_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8361
- Wer: 0.6474

## 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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 2.1771        | 4.62  | 300  | 0.9227          | 0.7302 |
| 0.56          | 9.23  | 600  | 0.8361          | 0.6474 |


### Framework versions

- Transformers 4.37.1
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1