File size: 2,350 Bytes
7ed135e
 
 
 
 
 
 
 
 
d74fc01
7ed135e
 
 
 
 
 
d74fc01
7ed135e
d74fc01
7ed135e
d74fc01
 
7ed135e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d74fc01
7ed135e
 
 
 
 
 
d74fc01
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7ed135e
 
 
 
 
 
d74fc01
7ed135e
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
---
language:
- hi
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
model-index:
- name: Whisper Small ko-Yfreq - syp1229
  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 ko-Yfreq - syp1229

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the aihub Y dialogue dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1951
- Cer: 0.0822

## 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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Cer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.2218        | 0.3   | 100  | 0.2452          | 0.0622 |
| 0.2261        | 0.59  | 200  | 0.2389          | 0.0653 |
| 0.2032        | 0.89  | 300  | 0.2079          | 0.0554 |
| 0.0821        | 1.18  | 400  | 0.2152          | 0.0630 |
| 0.0763        | 1.48  | 500  | 0.2025          | 0.0520 |
| 0.0778        | 1.78  | 600  | 0.1997          | 0.0611 |
| 0.0216        | 2.07  | 700  | 0.1918          | 0.0609 |
| 0.0226        | 2.37  | 800  | 0.1884          | 0.0663 |
| 0.0289        | 2.66  | 900  | 0.1940          | 0.2490 |
| 0.0212        | 2.96  | 1000 | 0.1925          | 0.1068 |
| 0.0075        | 3.25  | 1100 | 0.1966          | 0.0615 |
| 0.0044        | 3.55  | 1200 | 0.1986          | 0.0629 |
| 0.0051        | 3.85  | 1300 | 0.1950          | 0.1097 |
| 0.003         | 4.14  | 1400 | 0.1929          | 0.0689 |
| 0.0029        | 4.44  | 1500 | 0.1958          | 0.0774 |
| 0.0029        | 4.73  | 1600 | 0.1951          | 0.0822 |


### Framework versions

- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3