whisper-medium-nyagen-baseline-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: 0.4721
• Wer: 0.3136

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• 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
1.6573	1.0257	200	0.7472	0.4663
0.8271	2.0514	400	0.4985	0.4389
0.4472	3.0771	600	0.4721	0.3136
0.1499	4.1028	800	0.4968	0.3092
0.064	5.1285	1000	0.4892	0.3335
0.0521	6.1542	1200	0.5093	0.3048

Framework versions

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