Whisper for Turkish Call Centers
This model is a fine-tuned version of openai/whisper-large-v2 on the Custom turkish call center simulated data dataset. It achieves the following results on the evaluation set:
- Loss: 0.2338
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: 32
- 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_steps: 500
- training_steps: 2200
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3001 | 0.1777 | 250 | 0.2719 |
| 0.3084 | 0.3554 | 500 | 0.2729 |
| 0.2717 | 0.5330 | 750 | 0.2663 |
| 0.2561 | 0.7107 | 1000 | 0.2549 |
| 0.2564 | 0.8884 | 1250 | 0.2492 |
| 0.2067 | 1.0661 | 1500 | 0.2443 |
| 0.2023 | 1.2438 | 1750 | 0.2391 |
| 0.2222 | 1.4215 | 2000 | 0.2360 |
| 0.1928 | 1.5636 | 2200 | 0.2338 |
Framework versions
- Transformers 5.5.0
- Pytorch 2.10.0+cu128
- Datasets 4.8.4
- Tokenizers 0.22.2