ngia/whisper-small-pulaar

Whisper Small Pulaar

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

• Loss: 1.1167
• Wer: 120.9982

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: 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
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.9523	1.0373	1000	1.2430	132.9187
0.649	2.0747	2000	1.1308	113.6540
0.5278	3.1120	3000	1.1119	113.9935
0.4196	4.1494	4000	1.1167	120.9982

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2