whisper-tiny-zero-shot
This model is a fine-tuned version of openai/whisper-tiny on the Speech Commands dataset. It achieves the following results on the evaluation set:
- Loss: 2.3696
- Accuracy: 0.0
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: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- 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_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8182 | 1.0 | 6 | 1.7915 | 0.3333 |
| 1.7461 | 2.0 | 12 | 1.8821 | 0.0 |
| 1.6025 | 3.0 | 18 | 1.8774 | 0.0 |
| 1.4742 | 4.0 | 24 | 1.9225 | 0.0 |
| 1.1823 | 5.0 | 30 | 2.0195 | 0.0 |
| 0.9187 | 6.0 | 36 | 2.1978 | 0.0 |
| 0.9999 | 7.0 | 42 | 2.2030 | 0.0 |
| 0.6854 | 8.0 | 48 | 2.2368 | 0.0 |
| 0.8815 | 9.0 | 54 | 2.3313 | 0.0 |
| 0.5719 | 10.0 | 60 | 2.3696 | 0.0 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0