whisper-small-dioula
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: 1.2308
- Wer: 79.9333
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-06
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- 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: 100
- training_steps: 1500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.6770 | 0.9901 | 250 | 1.3940 | 85.7063 |
| 4.6025 | 1.9782 | 500 | 1.2605 | 84.8053 |
| 3.7246 | 2.9663 | 750 | 1.2230 | 84.4716 |
| 3.3528 | 3.9545 | 1000 | 1.2163 | 79.9555 |
| 3.0999 | 4.9426 | 1250 | 1.2202 | 80.5117 |
| 2.8991 | 5.9307 | 1500 | 1.2308 | 79.9333 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 4.8.3
- Tokenizers 0.22.2