Whisper Medium Galician
This model is a fine-tuned version of openai/whisper-medium on the mozilla-foundation/common_voice_13_0 gl dataset. It achieves the following results on the evaluation set:
- Loss: 0.3196
- Wer: 7.1227
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: 64
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 10000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0124 | 4.02 | 1000 | 0.2194 | 7.5383 |
| 0.0027 | 9.02 | 2000 | 0.2400 | 7.3382 |
| 0.0019 | 14.02 | 3000 | 0.2426 | 7.4055 |
| 0.0011 | 19.02 | 4000 | 0.2689 | 7.3520 |
| 0.0014 | 24.02 | 5000 | 0.2849 | 7.5314 |
| 0.0004 | 29.02 | 6000 | 0.2932 | 7.2589 |
| 0.0001 | 34.02 | 7000 | 0.3069 | 7.1485 |
| 0.0001 | 39.02 | 8000 | 0.3143 | 7.1485 |
| 0.0001 | 44.02 | 9000 | 0.3196 | 7.1227 |
| 0.0001 | 49.02 | 10000 | 0.3218 | 7.1244 |
Framework versions
- Transformers 4.33.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3
Citation
If you use these models in your research, please cite:
@misc{dezuazo2025whisperlmimprovingasrmodels,
title={Whisper-LM: Improving ASR Models with Language Models for Low-Resource Languages},
author={Xabier de Zuazo and Eva Navas and Ibon Saratxaga and Inma Hernáez Rioja},
year={2025},
eprint={2503.23542},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2503.23542},
}
Please, check the related paper preprint in arXiv:2503.23542 for more details.
Licensing
This model is available under the Apache-2.0 License. You are free to use, modify, and distribute this model as long as you credit the original creators.