whisper-medium-ln-ojpl-2

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

• Loss: 1.1202
• Wer Ortho: 35.8309
• Wer: 0.2901

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: 8
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 500
• training_steps: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.0172	23.19	1000	0.9966	41.9139	0.3407
0.0053	46.38	2000	1.0716	37.0920	0.2996
0.0034	69.57	3000	1.1329	36.0163	0.2850
0.0021	92.75	4000	1.1202	35.8309	0.2901

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 1.13.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3