lkhagvaa12/whisper-tiny-mn

Whisper tiny

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

• Loss: 0.4842
• Wer Ortho: 60.9132
• Wer: 60.9029

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: 32
• eval_batch_size: 32
• 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: constant_with_warmup
• lr_scheduler_warmup_steps: 50
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
1.0053	1.5198	500	0.6511	72.9011	72.8777
0.7337	3.0395	1000	0.5380	65.5752	65.5485
0.5557	4.5593	1500	0.4971	63.5066	63.4855
0.3969	6.0790	2000	0.4842	60.9132	60.9029

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2