liuh6/whisper-tiny_to_Russian_accent

Whisper tiny Korean

This model is a fine-tuned version of openai/whisper-tiny on the Korean English dataset. It achieves the following results on the evaluation set:

• Loss: 0.2310
• Wer: 10.1704

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: 2
• eval_batch_size: 1
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 1500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3176	0.5353	500	0.2917	13.9588
0.1752	1.0707	1000	0.2376	10.3483
0.1248	1.6060	1500	0.2310	10.1704

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.0
• Tokenizers 0.21.0