whisper-large-v2-medical-7
This model is a fine-tuned version of openai/whisper-large-v2 on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1337
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
No log | 0.26 | 20 | 0.5709 |
0.9523 | 0.51 | 40 | 0.3137 |
0.3351 | 0.77 | 60 | 0.1660 |
0.1566 | 1.03 | 80 | 0.1514 |
0.0977 | 1.28 | 100 | 0.1450 |
0.0977 | 1.54 | 120 | 0.1409 |
0.0899 | 1.79 | 140 | 0.1362 |
0.0922 | 2.05 | 160 | 0.1341 |
0.0644 | 2.31 | 180 | 0.1358 |
0.0544 | 2.56 | 200 | 0.1348 |
0.0544 | 2.82 | 220 | 0.1337 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.4.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 3
Model tree for tgrhn/whisper-large-v2-medical-7
Base model
openai/whisper-large-v2