whisper-medium-ft-GPT

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: 0.4126
• Wer: 16.1319
• Gpt: 8.1573

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	Gpt
1.4571	1.0	63	0.4337	33.9572	8.0230
0.176	2.0	126	0.3830	31.2834	7.9877
0.0563	3.0	189	0.3942	16.5775	8.1149
0.0192	4.0	252	0.4223	15.6863	7.9620
0.0079	5.0	315	0.4049	15.5080	7.6897
0.0023	6.0	378	0.4126	16.1319	8.1573

Framework versions

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