Whisper turbo KO - GTH
This model is a fine-tuned version of openai/whisper-large-v3-turbo on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5105
- Wer: 27.0053
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: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0087 | 22.2222 | 1000 | 0.4786 | 30.3667 |
| 0.0014 | 44.4444 | 2000 | 0.4935 | 30.7105 |
| 0.0001 | 66.6667 | 3000 | 0.5067 | 27.0817 |
| 0.0001 | 88.8889 | 4000 | 0.5105 | 27.0053 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3
- Downloads last month
- 2
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.