<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
Whisper Medium - Arabic language
This model is a fine-tuned version of MohammadJamalaldeen/whisper-medium-with-google-fleurs-ar-4000_steps on Google Fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.3975
- Wer: 17.5375
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0187 | 3.8 | 1000 | 0.3512 | 19.2 |
| 0.0017 | 7.6 | 2000 | 0.3857 | 18.0625 |
| 0.0002 | 11.41 | 3000 | 0.3894 | 17.5875 |
| 0.0001 | 15.21 | 4000 | 0.3975 | 17.5375 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu116
- Datasets 2.7.0
- Tokenizers 0.13.2
