sitwala/whisper-large-v3-anv-sot

Whisper whisper-large-v3 sot

This model is a fine-tuned version of openai/whisper-large-v3 on the dsfsi-anv/za-african-next-voices dataset. It achieves the following results on the evaluation set:

• Loss: 0.4477
• Wer: 22.0832

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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: constant_with_warmup
• lr_scheduler_warmup_steps: 100
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.9876	0.2	200	0.7042	44.9886
0.3854	0.4	400	0.5672	28.1945
0.2921	0.6	600	0.5097	24.3182
0.7161	0.8	800	0.4786	27.4673
0.5566	1.0	1000	0.4477	22.0832

Framework versions

• Transformers 4.52.0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.4