rahafvii/ABG_STT

arabic Whisper Small

This model is a fine-tuned version of openai/whisper-small on the Common Voice 13.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3384
• Wer: 44.4075

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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3476	0.4148	1000	0.4130	52.3435
0.2522	0.8295	2000	0.3676	49.2305
0.1606	1.2443	3000	0.3475	44.8855
0.161	1.6591	4000	0.3384	44.4075

Framework versions

• Transformers 4.44.2
• Pytorch 2.3.0+cu118
• Datasets 3.0.0
• Tokenizers 0.19.1