GogetaBlueMUI/whisper-medium-ur-v2

Whisper Medium Ur - Muhammad Abdullah

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

• Loss: 0.4708
• Wer: 28.3596

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: 7e-06
• train_batch_size: 32
• eval_batch_size: 16
• 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: 90
• training_steps: 800
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1373	0.5952	100	0.4251	27.3170
0.081	1.1905	200	0.4459	30.8355
0.0867	1.7857	300	0.4392	29.1326
0.0352	2.3810	400	0.4708	28.3596

Framework versions

• Transformers 4.48.2
• Pytorch 2.3.0+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0