whisper-sm-arabic-nouraa5

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: 0.7124
• Wer: 90.9278

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: 16
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• 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
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	1.0	32	3.4835	64.3918
2.7298	2.0	64	2.1102	97.8557
2.7298	3.0	96	1.4302	82.5361
1.2041	4.0	128	1.2875	73.7113
0.7692	5.0	160	1.1756	67.5464
0.7692	6.0	192	1.0568	67.5876
0.5691	7.0	224	0.7701	61.3196
0.2735	8.0	256	0.6364	79.8351
0.2735	9.0	288	0.6546	82.8247
0.0953	9.704	310	0.7124	90.9278

Framework versions

• Transformers 4.51.1
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1