itskavya/whisper-medium-informal-arabic

Whisper Medium Informal Arabic

This model is a fine-tuned version of openai/whisper-medium on the Informal Arabic dataset. It achieves the following results on the evaluation set:

• Loss: 0.5486
• Wer: 27.5165

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: 16
• 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: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0515	4.0650	1000	0.4619	30.4498
0.0028	8.1301	2000	0.5296	28.2894
0.0005	12.1951	3000	0.5486	27.5165

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0