metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-medium.en
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-medium.en
results: []
whisper-medium.en
This model is a fine-tuned version of openai/whisper-medium.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5138
- Wer: 0.1293
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: 5e-06
- train_batch_size: 32
- eval_batch_size: 4
- 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: 25
- training_steps: 300
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.1712 | 2.2727 | 25 | 1.4956 | 0.2397 |
| 0.8938 | 4.5455 | 50 | 0.9926 | 0.1682 |
| 0.5426 | 6.8182 | 75 | 0.8672 | 0.1535 |
| 0.4142 | 9.0909 | 100 | 0.7799 | 0.1462 |
| 0.3002 | 11.3636 | 125 | 0.6942 | 0.1409 |
| 0.2155 | 13.6364 | 150 | 0.6290 | 0.1346 |
| 0.1524 | 15.9091 | 175 | 0.5800 | 0.1335 |
| 0.1109 | 18.1818 | 200 | 0.5496 | 0.1314 |
| 0.0826 | 20.4545 | 225 | 0.5317 | 0.1304 |
| 0.0681 | 22.7273 | 250 | 0.5210 | 0.1304 |
| 0.0575 | 25.0 | 275 | 0.5160 | 0.1293 |
| 0.053 | 27.2727 | 300 | 0.5138 | 0.1293 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.6.0+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0