whisper-small-arabic-finetuned-on-halabi_daataset_no-diacritics-2

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2181
• Wer: 0.2491

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: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use 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: 200
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0426	3.5133	200	0.2065	0.2491
0.0069	7.0177	400	0.2383	0.2585
0.0021	10.5310	600	0.2496	0.2736
0.0007	14.0354	800	0.2582	0.2786
0.0006	17.5487	1000	0.2600	0.2765

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.5.1
• Datasets 3.1.0
• Tokenizers 0.20.3