File size: 1,704 Bytes
527368b
 
 
 
 
 
 
 
 
 
36e155c
5362366
527368b
 
 
 
 
36e155c
527368b
bda7975
527368b
e58f44e
6b391dc
527368b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6b391dc
527368b
 
6b391dc
 
527368b
 
bda7975
c8c8af5
527368b
 
 
6b391dc
 
 
 
 
 
527368b
 
 
 
e58f44e
527368b
e58f44e
bda7975
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
---
language:
- nan
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_1
model-index:
- name: Whisper Small Taiwanese
  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. -->

# Whisper Small Taiwanese

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 16.1 dataset.
It achieves the following results on the evaluation set:
- Cer: 48.6382
- Loss: 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: 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_ratio: 0.1
- num_epochs: 3

### Training results

| Training Loss | Epoch  | Step | Cer     | Validation Loss |
|:-------------:|:------:|:----:|:-------:|:---------------:|
| 0.9789        | 0.6452 | 1000 | 60.2169 | 0.9021          |
| 0.61          | 1.2903 | 2000 | 53.3884 | 0.7536          |
| 0.5611        | 1.9355 | 3000 | 51.3360 | 0.6703          |
| 0.3359        | 2.5806 | 4000 | 48.6382 | 0.6474          |


### Framework versions

- Transformers 4.40.1
- Pytorch 2.1.2
- Datasets 2.19.0
- Tokenizers 0.19.1