Abstract: While DCGAN as deep learning model utilizing spectrogram, allows for detection of deepfake audio, it is prone to overfitting which affects its ability to discriminate between real and fake ...

There was an error while loading. Please reload this page.

Abstract: Parameter-efficient transfer learning (PETL) methods have emerged as a solid alternative to the standard full fine-tuning approach. They only train a few extra parameters for each downstream ...

Music is an essential part of human culture, but automatically classifying songs into genres is a challenging problem for computers. With the explosion of digital music libraries, manual tagging is ...