Whisper Small af
This model is a fine-tuned version of openai/whisper-small on the Fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.8630
- Wer: 32.0866
- Cer: 11.9814
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: 32
- 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_ratio: 0.04
- training_steps: 600
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.3398 | 1.0583 | 100 | 0.8004 | 32.3290 | 11.4301 |
| 0.1013 | 3.0083 | 200 | 0.7973 | 31.9307 | 12.0576 |
| 0.0341 | 4.0667 | 300 | 0.8230 | 30.5281 | 11.0694 |
| 0.0136 | 6.0167 | 400 | 0.8477 | 32.8485 | 12.2775 |
| 0.0093 | 7.075 | 500 | 0.8606 | 33.5931 | 13.3537 |
| 0.0068 | 9.025 | 600 | 0.8630 | 32.0866 | 11.9814 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
Citation
Please cite the model using the following BibTeX entry:
@misc{deepdml/whisper-small-af-fleurs-norm,
title={Fine-tuned Whisper small ASR model for speech recognition in Afrikaans},
author={Jimenez, David},
howpublished={\url{https://huggingface.co/deepdml/whisper-small-af-fleurs-norm}},
year={2026}
}