whisper-large-v3-turbo-med-pl-lora-r64-enc-dec-lr2e-04-ep5-whisper_bigos10k_fair-filtered
This model is a fine-tuned version of openai/whisper-large-v3-turbo on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6342
- Model Preparation Time: 0.0204
- Wer: 13.1507
- Cer: 4.5137
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- 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_ratio: 0.1
- num_epochs: 5
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer | Cer |
|---|---|---|---|---|---|---|
| 1.6224 | 1.0 | 1780 | 1.6564 | 0.0204 | 13.8622 | 4.5215 |
| 1.5902 | 2.0 | 3560 | 1.6411 | 0.0204 | 13.3057 | 4.2441 |
| 1.5654 | 3.0 | 5340 | 1.6320 | 0.0204 | 12.8279 | 4.2279 |
| 1.5458 | 4.0 | 7120 | 1.6329 | 0.0204 | 13.1125 | 4.4243 |
| 1.5322 | 5.0 | 8900 | 1.6342 | 0.0204 | 13.1507 | 4.5137 |
Framework versions
- PEFT 0.18.1
- Transformers 4.57.6
- Pytorch 2.8.0+cu128
- Datasets 4.5.0
- Tokenizers 0.22.2