Whisper Large UAE
This model is a fine-tuned version of openai/whisper-Large on the high_quality_ds dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Wer: 0.0261
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: 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: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0024 | 15.8730 | 1000 | 0.0048 | 0.1303 |
| 0.0001 | 31.7460 | 2000 | 0.0001 | 0.0261 |
| 0.0 | 47.6190 | 3000 | 0.0001 | 0.0261 |
| 0.0 | 63.4921 | 4000 | 0.0000 | 0.0261 |
Framework versions
- Transformers 4.48.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0
- Downloads last month
- 7
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.