---
base_model: openai/whisper-medium
datasets:
- Marcusxx/chungnam_firestation
language:
- ko
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
model-index:
- name: chungnam_firestation_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. -->

# chungnam_firestation_model

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Marcusxx/chungnam_firestation dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2418
- Cer: 90.4973

## 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: 100
- training_steps: 5000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Cer     |
|:-------------:|:-------:|:----:|:---------------:|:-------:|
| 0.0211        | 9.8039  | 500  | 0.8797          | 88.9876 |
| 0.0024        | 19.6078 | 1000 | 0.9828          | 65.8970 |
| 0.0003        | 29.4118 | 1500 | 0.9937          | 42.6288 |
| 0.0002        | 39.2157 | 2000 | 1.0748          | 44.6714 |
| 0.0001        | 49.0196 | 2500 | 1.1177          | 56.2167 |
| 0.0001        | 58.8235 | 3000 | 1.1595          | 50.6217 |
| 0.0001        | 68.6275 | 3500 | 1.1969          | 62.8774 |
| 0.0001        | 78.4314 | 4000 | 1.2319          | 88.3659 |
| 0.0           | 88.2353 | 4500 | 1.2339          | 79.3073 |
| 0.0           | 98.0392 | 5000 | 1.2418          | 90.4973 |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.2.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1