whisper-hf-adalora-v2
This model is a fine-tuned version of openai/whisper-large-v3-turbo on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.7901
- Wer: 0.2574
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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- 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: constant_with_warmup
- lr_scheduler_warmup_steps: 500
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 13.472 | 1.0 | 92 | 3.2067 | 0.2570 |
| 13.6271 | 2.0 | 184 | 3.1891 | 0.2570 |
| 13.2512 | 3.0 | 276 | 3.1582 | 0.2570 |
| 13.4254 | 4.0 | 368 | 3.1123 | 0.2570 |
| 13.1894 | 5.0 | 460 | 3.0484 | 0.2570 |
| 12.8321 | 6.0 | 552 | 2.9636 | 0.2570 |
| 12.6423 | 7.0 | 644 | 2.8685 | 0.2572 |
| 12.3125 | 8.0 | 736 | 2.7901 | 0.2574 |
Framework versions
- PEFT 0.17.0
- Transformers 4.55.0
- Pytorch 2.6.0+cu124
- Datasets 2.18.0
- Tokenizers 0.21.2
- Downloads last month
- 4
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support