---
language:
- ko
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
base_model: openai/whisper-small
datasets:
- Marcusxx/gwanju
model-index:
- name: gwanju_small_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_small_model

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co./openai/whisper-small) on the Marcusxx/gwanju dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6780
- Cer: 250.8533

## 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: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 250
- training_steps: 10000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step  | Validation Loss | Cer      |
|:-------------:|:-------:|:-----:|:---------------:|:--------:|
| 0.4654        | 0.5928  | 500   | 0.4638          | 87.7337  |
| 0.333         | 1.1855  | 1000  | 0.4244          | 178.8377 |
| 0.3286        | 1.7783  | 1500  | 0.4122          | 177.5061 |
| 0.2259        | 2.3711  | 2000  | 0.4145          | 308.2453 |
| 0.2364        | 2.9638  | 2500  | 0.4093          | 108.4814 |
| 0.1643        | 3.5566  | 3000  | 0.4257          | 204.9992 |
| 0.1078        | 4.1494  | 3500  | 0.4426          | 293.1846 |
| 0.111         | 4.7421  | 4000  | 0.4517          | 169.9893 |
| 0.0702        | 5.3349  | 4500  | 0.4771          | 297.9658 |
| 0.064         | 5.9277  | 5000  | 0.4881          | 245.7345 |
| 0.0417        | 6.5205  | 5500  | 0.5119          | 193.3111 |
| 0.024         | 7.1132  | 6000  | 0.5500          | 282.1528 |
| 0.0278        | 7.7060  | 6500  | 0.5631          | 188.2193 |
| 0.014         | 8.2988  | 7000  | 0.6062          | 257.3118 |
| 0.0164        | 8.8915  | 7500  | 0.6047          | 235.9248 |
| 0.0088        | 9.4843  | 8000  | 0.6341          | 234.2000 |
| 0.0061        | 10.0771 | 8500  | 0.6508          | 239.4206 |
| 0.0064        | 10.6698 | 9000  | 0.6657          | 252.8322 |
| 0.0039        | 11.2626 | 9500  | 0.6747          | 250.7573 |
| 0.0044        | 11.8554 | 10000 | 0.6780          | 250.8533 |


### Framework versions

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