ybendou/arabic-whisper-large-v3-turbo-v2

Whisper large-v3-turbo Ar

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1256
• Model Preparation Time: 0.024
• Wer: 34.2044
• Cer: 17.8321

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: 4.375e-06
• train_batch_size: 2
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 16
• 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: 10
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Wer	Cer
0.1838	1.0	5625	0.1547	0.024	45.8608	24.5044
0.0959	2.0	11250	0.1285	0.024	38.6960	20.2074
0.0863	3.0	16875	0.1209	0.024	36.6832	19.1926
0.0586	4.0	22500	0.1179	0.024	34.9463	18.3073
0.0594	5.0	28125	0.1256	0.024	34.2044	17.8321
0.0181	6.0	33750	0.1342	0.024	34.4860	17.9863

Framework versions

• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0