metadata
language:
- ko
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- kresnik/zeroth_korean
metrics:
- wer
model-index:
- name: Whisper Small Ko - haseong8012
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: kresnik/zeroth_korean
type: kresnik/zeroth_korean
config: clean
split: test
args: 'config: ko, split: test'
metrics:
- name: Wer
type: wer
value: 9.351001355217587
Whisper Small Ko - haseong8012
This model is a fine-tuned version of openai/whisper-small on the kresnik/zeroth_korean dataset. It achieves the following results on the evaluation set:
- Loss: 0.0965
- Wer: 9.3510
- Cer: 4.1693
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: 5e-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: 4000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.1738 | 0.72 | 1000 | 0.2348 | 20.1777 | 8.0402 |
| 0.0601 | 1.44 | 2000 | 0.1447 | 16.1873 | 7.3218 |
| 0.0148 | 2.16 | 3000 | 0.1103 | 15.1784 | 7.6162 |
| 0.0155 | 2.87 | 4000 | 0.0965 | 9.3510 | 4.1693 |
Framework versions
- Transformers 4.33.2
- Pytorch 1.12.1
- Datasets 2.14.5
- Tokenizers 0.13.3