Whisper FR RAD
Collection
Ensemble de Modèles issues du Fine Tuning de Whisper sur un petit ensemble d'audios de rapports de radiologie • 4 items • Updated
Whisper Tiny Fr - Radiologie
This model is a fine-tuned version of openai/whisper-tiny on 107 collected audios based on medical radiological report corpora.
Model description
Whisper Tiny Fr - Radiology
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: 3e-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: 5000
- mixed_precision_training: Native AMP
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
- Downloads last month
- 6
Model tree for StephaneBah/whisper-tiny-rad-fr
Base model
openai/whisper-tiny