Whisper Small Lingala
This model is a fine-tuned version of openai/whisper-small on the FLEURS dataset. It achieves the following results on the evaluation set:
- Loss: 0.5608
- Wer: 19.8362
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 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: 500
- training_steps: 3000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0939 | 4.7619 | 1000 | 0.4101 | 20.1182 |
| 0.0034 | 9.5238 | 2000 | 0.4968 | 19.7772 |
| 0.0013 | 14.2857 | 3000 | 0.5139 | 19.3226 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0