whisper-small-yoruba-dora-st
This model is a fine-tuned version of openai/whisper-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1896
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0684 | 0.4343 | 1000 | 0.2259 |
| 0.0712 | 0.8685 | 2000 | 0.2108 |
| 0.0355 | 1.3027 | 3000 | 0.2083 |
| 0.0353 | 1.7369 | 4000 | 0.1946 |
| 0.0178 | 2.1711 | 5000 | 0.1944 |
| 0.0147 | 2.6054 | 6000 | 0.1896 |
Framework versions
- PEFT 0.14.0
- Transformers 4.50.3
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.21.4