pashto-voice-v2

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: 0.5255
• Wer: 0.2770

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 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
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5178	0.9999	4687	0.2928	0.3004
0.3795	1.9997	9374	0.2649	0.2722
0.2977	2.9995	14061	0.2630	0.2752
0.2195	3.9993	18748	0.2750	0.2650
0.1530	4.9990	23435	0.3107	0.2710
0.0920	5.9988	28122	0.3485	0.2668
0.0606	6.9986	32809	0.4061	0.2776
0.0290	7.9984	37496	0.4569	0.2710
0.0122	8.9982	42183	0.5061	0.2794
0.0072	9.9980	46870	0.5256	0.2764
0.0087	10.0	46880	0.5255	0.2770

Framework versions

• Transformers 5.13.0
• Pytorch 2.6.0+cu124
• Datasets 5.0.0
• Tokenizers 0.22.2