whisper-medium-gabriel_fold_0

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

• Loss: 0.2805
• Wer: 11.1420

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.7317	1.0	169	0.2706	12.8086
0.1187	2.0	338	0.2537	12.0062
0.0431	3.0	507	0.2580	11.3272
0.0161	4.0	676	0.2716	11.3580
0.0048	5.0	845	0.2804	11.2963
0.0018	6.0	1014	0.2805	11.1420

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3