deepdml/whisper-tiny-af-fleurs-norm

Whisper Tiny af

This model is a fine-tuned version of openai/whisper-tiny on the Fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 1.2862
• Wer: 49.5238
• Cer: 21.6345

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: 64
• 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
0.9449	3.0017	100	1.3743	55.2381	21.4644
0.4604	6.0033	200	1.2738	51.1515	20.4000
0.2774	9.005	300	1.2614	49.8528	20.9014
0.1969	12.0067	400	1.2717	50.2857	21.6315
0.1467	15.0083	500	1.2801	49.7662	21.5113
0.1375	18.01	600	1.2862	49.5238	21.6345

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-tiny-af-fleurs-norm,
      title={Fine-tuned Whisper tiny ASR model for speech recognition in Afrikaans},
      author={Jimenez, David},
      howpublished={\url{https://huggingface.co/deepdml/whisper-tiny-af-fleurs-norm}},
      year={2026}
    }