Whisper medium Bengali
This model is a fine-tuned version of bengaliAI/tugstugi_bengaliai-regional-asr_whisper-medium on the Bengali ASR Chunked dataset. It achieves the following results on the evaluation set:
- Loss: 0.1587
- Wer: 23.3549
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- 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: 200
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.9011 | 0.9974 | 190 | 0.1874 | 29.3577 |
| 0.5526 | 1.9921 | 380 | 0.1597 | 23.6563 |
| 0.3047 | 2.9869 | 570 | 0.1587 | 23.3549 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2