| --- |
| library_name: transformers |
| license: apache-2.0 |
| base_model: sulaimank/whisper-small-CV-Fleurs-lg-300 |
| tags: |
| - generated_from_trainer |
| metrics: |
| - wer |
| model-index: |
| - name: whisper-small-lug-4src |
| 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-lug-4src |
|
|
| This model is a fine-tuned version of [sulaimank/whisper-small-CV-Fleurs-lg-300](https://huggingface.co/sulaimank/whisper-small-CV-Fleurs-lg-300) on the None dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.6772 |
| - Wer: 0.2145 |
| - Cer: 0.1136 |
|
|
| ## 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: 8 |
| - eval_batch_size: 4 |
| - 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: 500 |
| - num_epochs: 30 |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Cer | Validation Loss | Wer | |
| |:-------------:|:-------:|:-----:|:------:|:---------------:|:------:| |
| | 0.6768 | 1.1508 | 1000 | 0.2882 | 0.4067 | 0.3768 | |
| | 0.3794 | 2.3017 | 2000 | 0.2152 | 0.3882 | 0.3322 | |
| | 0.3018 | 3.4525 | 3000 | 0.4216 | 0.2716 | 0.1417 | |
| | 0.2216 | 4.6033 | 4000 | 0.4807 | 0.2945 | 0.1581 | |
| | 0.1689 | 5.7542 | 5000 | 0.5392 | 0.2358 | 0.1309 | |
| | 0.0748 | 6.9050 | 6000 | 0.5808 | 0.2331 | 0.1226 | |
| | 0.0453 | 8.0553 | 7000 | 0.6295 | 0.2382 | 0.1290 | |
| | 0.0276 | 9.2061 | 8000 | 0.6217 | 0.2413 | 0.1291 | |
| | 0.0166 | 10.3569 | 9000 | 0.6351 | 0.2300 | 0.1183 | |
| | 0.0115 | 11.5078 | 10000 | 0.6329 | 0.2366 | 0.1237 | |
| | 0.0096 | 12.6586 | 11000 | 0.6458 | 0.2428 | 0.1259 | |
| | 0.0059 | 13.8094 | 12000 | 0.6414 | 0.2323 | 0.1182 | |
| | 0.0045 | 14.9603 | 13000 | 0.6418 | 0.2638 | 0.1463 | |
| | 0.0043 | 16.1105 | 14000 | 0.6474 | 0.2362 | 0.1269 | |
| | 0.0026 | 17.2614 | 15000 | 0.6542 | 0.2284 | 0.1259 | |
| | 0.0026 | 18.4122 | 16000 | 0.6498 | 0.2253 | 0.1164 | |
| | 0.0013 | 19.5630 | 17000 | 0.6504 | 0.2758 | 0.1572 | |
| | 0.0015 | 20.7139 | 18000 | 0.6608 | 0.2187 | 0.1161 | |
| | 0.0015 | 21.8647 | 19000 | 0.6602 | 0.2234 | 0.1173 | |
| | 0.0015 | 23.0150 | 20000 | 0.6661 | 0.2455 | 0.1296 | |
| | 0.0008 | 24.1658 | 21000 | 0.6625 | 0.2265 | 0.1214 | |
| | 0.0003 | 25.3166 | 22000 | 0.6698 | 0.2498 | 0.1384 | |
| | 0.0003 | 26.4675 | 23000 | 0.6725 | 0.2347 | 0.1234 | |
| | 0.0001 | 27.6183 | 24000 | 0.6738 | 0.2187 | 0.1194 | |
| | 0.0001 | 28.7691 | 25000 | 0.6766 | 0.2133 | 0.1144 | |
| | 0.0001 | 29.9200 | 26000 | 0.6772 | 0.2145 | 0.1136 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 5.3.0 |
| - Pytorch 2.10.0+cu130 |
| - Datasets 4.0.0 |
| - Tokenizers 0.22.2 |
| |
