nurzhanit/whisper-omg-2.5.4

Whisper Small Hi - Sanchit Gandhi

This model is a fine-tuned version of nurzhanit/whisper-enhanced-ml on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0019
• Wer: 0.5728

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 25
• training_steps: 250

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0722	1.4706	50	0.0438	6.0586
0.0318	2.9412	100	0.0132	2.1370
0.0098	4.4118	150	0.0051	1.0355
0.0058	5.8824	200	0.0025	0.6389
0.0018	7.3529	250	0.0019	0.5728

Framework versions

• Transformers 4.40.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.2
• Tokenizers 0.19.1