File size: 1,696 Bytes
6a4252a
5b7efd6
 
6a4252a
 
5b7efd6
6a4252a
0e21b47
5b7efd6
 
6a4252a
 
 
 
 
 
 
 
 
 
5b7efd6
6a4252a
91782ae
 
6a4252a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
91782ae
 
6a4252a
 
 
 
 
 
91782ae
 
 
 
6a4252a
 
 
 
 
 
 
 
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:
- ko
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
base_model: openai/whisper-large
datasets:
- Marcusxx/gwanju
model-index:
- name: gwanju_large_model
  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. -->

# gwanju_large_model

This model is a fine-tuned version of [openai/whisper-large](https://huggingface.co/openai/whisper-large) on the Marcusxx/gwanju dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5238
- Cer: 238.8739

## 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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 20000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step  | Validation Loss | Cer      |
|:-------------:|:------:|:-----:|:---------------:|:--------:|
| 0.2792        | 1.4819 | 5000  | 0.3803          | 453.3834 |
| 0.1361        | 2.9638 | 10000 | 0.3823          | 336.9865 |
| 0.0335        | 4.4458 | 15000 | 0.4644          | 344.2898 |
| 0.0097        | 5.9277 | 20000 | 0.5238          | 238.8739 |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.2.2+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1