doongsae/whisper_finetuning

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Fine-tuning whisper

This model is a fine-tuned version of openai/whisper-base on the TTS dataset for Capstone Design II project in Sogang University.

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 1e-05
• —trainbatchsize: 16
• —evalbatchsize: 8
• —seed: 42
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 10
• —training_steps: 100
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.36.0.dev0
• —Pytorch 2.1.0+cu118
• —Datasets 2.15.0
• —Tokenizers 0.15.0