whisper-large-v2-ft-ca-25awr

This model is a fine-tuned version of openai/whisper-large-v2 on the cymen-arfor/25awr default dataset. It achieves the following results on the evaluation set:

• Loss: 0.8440
• Wer: 0.4025

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: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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
0.4138	1.6488	1000	0.5216	0.5082
0.1535	3.2976	2000	0.5362	0.4263
0.084	4.9464	3000	0.5920	0.4038
0.0185	6.5952	4000	0.7443	0.4076
0.0038	8.2440	5000	0.8440	0.4025

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3