ciderstt/whisper-small-nan-tw

Whisper Small taiwanese

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 1.2406
• Wer: 103.9374

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: 8
• seed: 42
• optimizer: Use OptimizerNames.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_steps: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0007	39.005	1000	1.0996	97.8682
0.0003	79.005	2000	1.1532	100.1958
0.0001	119.005	3000	1.1976	102.4146
0.0001	159.005	4000	1.2206	105.9822
0.0001	199.005	5000	1.2406	103.9374

Framework versions

• Transformers 4.50.0.dev0
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0