Whisper Base Ta with Phone Control Data - Navin Kumar J
This model is a fine-tuned version of navin-kumar-j/whisper-base-ta on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0045
- Wer: 0.0077
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 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: 10
- training_steps: 200
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0007 | 1.25 | 40 | 0.0046 | 0.0058 |
| 0.0013 | 2.5 | 80 | 0.0045 | 0.0077 |
| 0.0018 | 3.75 | 120 | 0.0043 | 0.0077 |
| 0.0001 | 5.0 | 160 | 0.0048 | 0.0077 |
| 0.0001 | 6.25 | 200 | 0.0045 | 0.0077 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0+cu126
- Datasets 3.5.0
- Tokenizers 0.21.1
- Downloads last month
- -