alishahzad/whisper-small-hi

Whisper Small UR - Kissan Konnect

This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.8877
• Wer: 38.6506

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use 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: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2061	2.9806	1000	0.5796	37.1996
0.0302	5.9627	2000	0.6702	43.7793
0.0049	8.9478	3000	0.8019	35.7425
0.0012	11.9240	4000	0.8650	37.7654
0.0004	14.9001	5000	0.8877	38.6506

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.21.0