whisper-tiny-minds14-en-us
This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4819
- Wer Ortho: 0.3350
- Wer: 0.3282
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: 16
- 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: 40
- training_steps: 300
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.9129 | 1.7857 | 50 | 0.5355 | 0.3899 | 0.3808 |
| 0.2583 | 3.5714 | 100 | 0.4819 | 0.3350 | 0.3282 |
| 0.1079 | 5.3571 | 150 | 0.5057 | 0.3374 | 0.3288 |
| 0.0488 | 7.1429 | 200 | 0.5284 | 0.3498 | 0.3406 |
| 0.0230 | 8.9286 | 250 | 0.5447 | 0.3448 | 0.3365 |
| 0.0157 | 10.7143 | 300 | 0.5532 | 0.3448 | 0.3365 |
Framework versions
- Transformers 5.20.2
- Pytorch 2.11.0+cu130
- Datasets 3.6.0
- Tokenizers 0.23.3