deepdml/whisper-tiny-af-mix-norm

Whisper Tiny af

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• Loss: 1.6355
• Wer: 48.7273
• Cer: 19.0658

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: 128
• eval_batch_size: 128
• 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: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.0133	14.0068	1000	1.3968	48.6926	20.3824
0.0032	28.0136	2000	1.5069	48.6234	19.1039
0.0016	43.0066	3000	1.5777	48.6926	19.0775
0.0011	57.0134	4000	1.6203	48.4675	19.1273
0.0009	72.0064	5000	1.6355	48.7273	19.0658

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-mix-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-mix-norm}},
      year={2026}
    }