Whisper - Serbian Model
This model is a fine-tuned version of openai/whisper-medium on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4742
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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- 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: 50
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0953 | 4.6189 | 1000 | 0.3313 |
| 0.0237 | 9.2379 | 2000 | 0.3623 |
| 0.0019 | 13.8568 | 3000 | 0.4403 |
| 0.0007 | 18.4758 | 4000 | 0.4742 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.3
- Pytorch 2.5.1
- Datasets 3.0.0
- Tokenizers 0.20.3