zowlex/whisper-small-ar-common-17

Whisper Small Arabic

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

• Loss: 0.3517
• Wer: 44.3961

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2707	0.4119	1000	0.4188	51.3000
0.2452	0.8237	2000	0.3639	46.1863
0.1613	1.2356	3000	0.3470	44.9194
0.1382	1.6474	4000	0.3398	45.0351
0.1177	2.0593	5000	0.3502	44.5154
0.1206	2.4712	6000	0.3501	44.9781
0.1216	2.8830	7000	0.3423	43.5258
0.072	3.2949	8000	0.3517	44.3961

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0