Whisper medium FR
This model is a fine-tuned version of openai/whisper-medium on the CAENNAIS dataset. It achieves the following results on the evaluation set:
- eval_loss: 0.5070
- eval_wer: 30.7393
- eval_runtime: 137.6146
- eval_samples_per_second: 2.354
- eval_steps_per_second: 0.298
- epoch: 3.0
- step: 243
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0
- Downloads last month
- 4
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for Rziane/whisper-medium-fr
Base model
openai/whisper-medium