README.md

metadata

library_name: transformers
language:
  - ko
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - didiudom94/gentlemen
metrics:
  - wer
model-index:
  - name: Whisper Small-V2 Ko to En
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Gentlemen
          type: didiudom94/gentlemen
          args: 'split: train'
        metrics:
          - name: Wer
            type: wer
            value: 0.8733030859529726

Whisper Small-V2 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: 1.3724
• Wer: 0.8733

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-06
• 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: 400
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.4028	0.2253	1000	1.5128	1.0477
1.4054	0.4507	2000	1.4252	0.9156
1.3733	0.6760	3000	1.3890	0.9197
1.4428	0.9013	4000	1.3724	0.8733

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3