A LoRA-trained Whisper-Large-v3-turbo specifically in Yue-Chinese (lang_code: zh)
This model is a fine-tuned version of openai/whisper-large-v3-turbo on the Yue Dataset with Hong Kong style dataset. It achieves the following results on the evaluation set:
- Loss: 0.3416
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 50
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4077 | 1.0 | 49 | 0.3764 |
| 0.3357 | 2.0 | 98 | 0.3475 |
| 0.2779 | 3.0 | 147 | 0.3416 |
Framework versions
- PEFT 0.18.0
- Transformers 4.57.2
- Pytorch 2.9.0+cu126
- Datasets 4.4.1
- Tokenizers 0.22.1