metadata
library_name: transformers
language:
- ko
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- didiudom94/gentlemen
metrics:
- bleu
model-index:
- name: Whisper Small Ko to En
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Gentlemen
type: didiudom94/gentlemen
args: 'split: train'
metrics:
- name: Bleu
type: bleu
value: 0.1392438982977928
Whisper Small Ko to En
This model is a fine-tuned version of openai/whisper-small on the Gentlemen dataset. It achieves the following results on the evaluation set:
- Loss: 1.3270
- Bleu: 0.1392
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: 8
- 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu |
|---|---|---|---|---|
| 0.8182 | 0.2253 | 1000 | 1.6561 | 0.1004 |
| 1.4212 | 0.4507 | 2000 | 1.4204 | 0.1195 |
| 1.3578 | 0.6760 | 3000 | 1.3638 | 0.1320 |
| 1.3446 | 0.9013 | 4000 | 1.3265 | 0.1356 |
| 0.9391 | 1.1266 | 5000 | 1.3270 | 0.1392 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.20.3