Shubhr Singh

Profile

Project Title:  

Advancing Audio Classification with Graph Neural Networks

Abstract:

This project focuses on the innovative use of graph neural networks (GNNs) for audio-based applications, particularly in the realms of classification and tagging. By integrating GNNs, this research explores an alternative approach to conventional deep learning techniques, leveraging the inherent strengths of graph-based data processing.

The primary objective is to harness GNNs to incorporate structured prior knowledge, thereby facilitating more sophisticated and robust representation learning of audio data.

Teaching

Teaching Fellow – Music Informatics (postgraduate)

Demonstrator – Artificial Intelligence, Deep learning for audio and music

Research