Whisper Small Hindi Experimental
This model is a fine-tuned version of openai/whisper-small on the Common Voice 22.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3003
- Wer: 37.3543
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: 16
- eval_batch_size: 8
- seed: 42
- 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
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.3310 | 0.2110 | 100 | 0.4101 | 45.7002 |
| 0.2631 | 0.4219 | 200 | 0.3488 | 40.9286 |
| 0.2371 | 0.6329 | 300 | 0.3252 | 39.1379 |
| 0.2116 | 0.8439 | 400 | 0.3058 | 38.1371 |
| 0.1573 | 1.0549 | 500 | 0.3003 | 37.3543 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 3.6.0
- Tokenizers 0.22.2