TakHemlata / SSL_Anti-spoofing

This repository includes the code to reproduce our paper "Automatic speaker verification spoofing and deepfake detection using wav2vec 2.0 and data augmentation".

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

How to find the best model?

xmz1009 opened this issue · comments

Hello author,

may I ask that the model of each round of training has been saved in your code, because no verification indicators are given, then how to know that the model of that round of training has the best performance?

Thank you for your sharing!