metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: ap-dHsT9h4tktkDaOuJtOWql8
results: []
ap-dHsT9h4tktkDaOuJtOWql8
This model is a fine-tuned version of openai/whisper-large-v3 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3711
- Model Preparation Time: 0.0225
- Wer: 0.1160
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- 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: 400
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer |
|---|---|---|---|---|---|
| 0.3495 | 0.9791 | 41 | 0.3525 | 0.0225 | 0.1237 |
| 0.2341 | 1.9791 | 82 | 0.2712 | 0.0225 | 0.1079 |
| 0.1627 | 2.9791 | 123 | 0.2690 | 0.0225 | 0.1042 |
| 0.0835 | 3.9791 | 164 | 0.2909 | 0.0225 | 0.1058 |
| 0.0575 | 4.9791 | 205 | 0.3031 | 0.0225 | 0.1218 |
| 0.0388 | 5.9791 | 246 | 0.3359 | 0.0225 | 0.1098 |
| 0.0277 | 6.9791 | 287 | 0.3808 | 0.0225 | 0.1072 |
| 0.0203 | 7.9791 | 328 | 0.4040 | 0.0225 | 0.1059 |
| 0.0263 | 8.9791 | 369 | 0.3793 | 0.0225 | 0.1184 |
| 0.0253 | 9.9791 | 410 | 0.3711 | 0.0225 | 0.1160 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0