README.md

metadata

library_name: transformers
language:
  - ko
license: apache-2.0
base_model: openai/whisper-large-v2
tags:
  - generated_from_trainer
datasets:
  - didiudom94/gentlemen
metrics:
  - wer
model-index:
  - name: Whisper Large-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: 0.7335058679750543

Whisper Large-V2 Ko to En

This model is a fine-tuned version of openai/whisper-large-v2 on the Gentlemen dataset. It achieves the following results on the evaluation set:

• Loss: 1.0783
• Wer: 0.7335

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: 400
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.2034	0.2253	1000	1.2714	0.7977
1.1758	0.4507	2000	1.1820	0.7810
1.1567	0.6760	3000	1.1158	0.7590
1.0885	0.9013	4000	1.0783	0.7335

Framework versions

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