whisper-small-arbyeg_three

This model is a fine-tuned version of EYEDOL/whisper-small-arbyeg_TWO on the arabic_mozilla_dataset_clean dataset. It achieves the following results on the evaluation set:

• Loss: 0.2749
• Wer: 45.5142

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: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1606	0.7205	1000	0.3004	46.5663
0.1562	1.4409	2000	0.2749	45.5142

Framework versions

• Transformers 5.3.0
• Pytorch 2.9.0+cu126
• Datasets 4.8.3
• Tokenizers 0.22.2