| --- |
| library_name: peft |
| license: apache-2.0 |
| base_model: openai/whisper-small |
| tags: |
| - base_model:adapter:openai/whisper-small |
| - lora |
| - transformers |
| model-index: |
| - name: whisper-small-gui |
| results: [] |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # whisper-small-gui |
|
|
| This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.5692 |
| - Cer: 0.9526 |
|
|
| ## 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: 0.0003 |
| - train_batch_size: 8 |
| - eval_batch_size: 8 |
| - seed: 42 |
| - gradient_accumulation_steps: 2 |
| - total_train_batch_size: 16 |
| - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 100 |
| - num_epochs: 10 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Cer | |
| |:-------------:|:------:|:----:|:---------------:|:------:| |
| | 6.4677 | 0.4329 | 100 | 1.4408 | 0.9623 | |
| | 2.3434 | 0.8658 | 200 | 0.9404 | 0.8162 | |
| | 1.6431 | 1.2987 | 300 | 0.8061 | 0.8386 | |
| | 1.3529 | 1.7316 | 400 | 0.7035 | 0.9646 | |
| | 1.1665 | 2.1645 | 500 | 0.6440 | 0.9493 | |
| | 1.0026 | 2.5974 | 600 | 0.6067 | 0.9547 | |
| | 0.9371 | 3.0303 | 700 | 0.5685 | 0.9537 | |
| | 0.7621 | 3.4632 | 800 | 0.5687 | 0.8942 | |
| | 0.7186 | 3.8961 | 900 | 0.5301 | 0.9228 | |
| | 0.6113 | 4.3290 | 1000 | 0.5389 | 1.0200 | |
| | 0.5656 | 4.7619 | 1100 | 0.5438 | 0.9263 | |
| | 0.5255 | 5.1948 | 1200 | 0.5349 | 0.9130 | |
| | 0.4546 | 5.6277 | 1300 | 0.5307 | 1.0552 | |
| | 0.4159 | 6.0606 | 1400 | 0.5384 | 1.0177 | |
| | 0.3525 | 6.4935 | 1500 | 0.5420 | 1.0315 | |
| | 0.3394 | 6.9264 | 1600 | 0.5493 | 0.9588 | |
| | 0.2660 | 7.3593 | 1700 | 0.5604 | 0.9960 | |
| | 0.2773 | 7.7922 | 1800 | 0.5540 | 0.9739 | |
| | 0.2719 | 8.2251 | 1900 | 0.5683 | 0.9521 | |
| | 0.2068 | 8.6580 | 2000 | 0.5605 | 1.0251 | |
| | 0.2027 | 9.0909 | 2100 | 0.5644 | 0.9697 | |
| | 0.1911 | 9.5238 | 2200 | 0.5749 | 0.9335 | |
| | 0.1723 | 9.9567 | 2300 | 0.5692 | 0.9526 | |
| |
| |
| ### Framework versions |
| |
| - PEFT 0.18.1 |
| - Transformers 5.1.0 |
| - Pytorch 2.9.1+cu128 |
| - Datasets 3.6.0 |
| - Tokenizers 0.22.2 |
