Whisper medium pt jwlang - Michel Mesquita
This model is a fine-tuned version of openai/whisper-medium on the jwlang 1.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6361
- Wer: 18.7271
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: 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 |
0.0038 |
14.0845 |
1000 |
0.5291 |
20.3182 |
0.0001 |
28.1690 |
2000 |
0.6034 |
18.9718 |
0.0 |
42.2535 |
3000 |
0.6277 |
19.0942 |
0.0 |
56.3380 |
4000 |
0.6361 |
18.7271 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1