Whisper Tiny ta

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: 0.2641
Wer: 51.6662
Cer: 11.8757

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.04
training_steps: 8000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.2508	0.125	1000	0.3392	62.4235	15.5864
0.1747	0.25	2000	0.3003	57.3701	13.4963
0.1586	0.375	3000	0.2905	55.5023	13.1754
0.1244	0.5	4000	0.2812	53.6500	12.6062
0.1361	0.625	5000	0.2687	52.9080	12.3268
0.1093	0.75	6000	0.2685	52.2523	12.0787
0.1141	0.875	7000	0.2647	51.9844	11.9065
0.1274	1.0	8000	0.2641	51.6662	11.8757

Framework versions

Transformers 4.48.0.dev0
Pytorch 2.5.1+cu121
Datasets 3.6.0
Tokenizers 0.21.0

Citation

Please cite the model using the following BibTeX entry:

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