whisper-medium-nyagen-male-model

This model is a fine-tuned version of openai/whisper-medium on the nyagen dataset. It achieves the following results on the evaluation set:

• Loss: 0.2401
• Wer: 0.2025

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• 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
0.8171	0.9685	200	0.3874	0.2875
0.4046	1.9346	400	0.2777	0.2068
0.3008	2.9007	600	0.2401	0.2025
0.1009	3.8668	800	0.2750	0.1806
0.0663	4.8329	1000	0.2580	0.1687
0.0492	5.7990	1200	0.2699	0.1910

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0