whisper-medium-nyagen-female-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: 0.3796
• Wer: 0.2574

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
1.1709	1.1834	200	0.6065	0.3997
0.4797	2.3669	400	0.4227	0.3908
0.2388	3.5503	600	0.3936	0.2932
0.1376	4.7337	800	0.3796	0.2574
0.0765	5.9172	1000	0.4010	0.2842
0.0344	7.1006	1200	0.4034	0.2381
0.0207	8.2840	1400	0.4384	0.2314

Framework versions

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