pashto-asr-v6

This model is a fine-tuned version of ihanif/pashto-asr-v5 on the common_voice_19_0 dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.5639
• eval_wer: 30.1268
• eval_runtime: 85.7746
• eval_samples_per_second: 18.793
• eval_steps_per_second: 2.355
• epoch: 5.2185
• step: 1000

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 3000
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.19.1