File size: 2,499 Bytes
d359860
75e36c6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d359860
 
75e36c6
 
d359860
75e36c6
d359860
75e36c6
 
 
 
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
d359860
75e36c6
 
 
 
 
 
 
 
 
 
 
 
d359860
75e36c6
d359860
75e36c6
 
 
 
 
 
 
 
 
 
 
 
 
 
d359860
 
75e36c6
d359860
75e36c6
 
 
 
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
89
90
91
---
license: mit
base_model: facebook/w2v-bert-2.0
tags:
- generated_from_trainer
datasets:
- ml-superb-subset
metrics:
- wer
model-index:
- name: w2v-bert-2.0-ml-superb-xty
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: ml-superb-subset
      type: ml-superb-subset
      config: xty
      split: test
      args: xty
    metrics:
    - name: Wer
      type: wer
      value: 1.3984915147705845
---

<!-- 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-2.0-ml-superb-xty

This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the ml-superb-subset dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3981
- Wer: 1.3985

## 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: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- 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: linear
- lr_scheduler_warmup_steps: 30
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| 3.5467        | 0.8219 | 30   | 2.8636          | 1.0    |
| 2.4639        | 1.6438 | 60   | 2.5298          | 1.0094 |
| 2.38          | 2.4658 | 90   | 2.4983          | 1.1263 |
| 2.2725        | 3.2877 | 120  | 2.4866          | 1.2319 |
| 2.2608        | 4.1096 | 150  | 2.5116          | 1.5405 |
| 2.2222        | 4.9315 | 180  | 2.4588          | 1.3300 |
| 2.2609        | 5.7534 | 210  | 2.4448          | 1.3451 |
| 2.1665        | 6.5753 | 240  | 2.4270          | 1.3199 |
| 2.1703        | 7.3973 | 270  | 2.4223          | 1.3576 |
| 2.1366        | 8.2192 | 300  | 2.4054          | 1.4085 |
| 2.123         | 9.0411 | 330  | 2.4006          | 1.4180 |
| 2.1331        | 9.8630 | 360  | 2.3981          | 1.3985 |


### Framework versions

- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1