Audio Event-Relational Graph Representation Learning for Acoustic Scene Classification

TL;DR

This paper introduces a novel event-relational graph framework for acoustic scene classification that improves interpretability and achieves competitive performance by modeling relationships between audio events.

Contribution

It proposes the first event-relational graph learning framework for ASC, revealing cues used in scene classification and focusing on relationships between acoustic events.

Findings

ERGL achieves competitive accuracy on ASC datasets.

The approach effectively models relationships between audio events.

Visualizations demonstrate interpretability of the learned graphs.

Abstract

Most deep learning-based acoustic scene classification (ASC) approaches identify scenes based on acoustic features converted from audio clips containing mixed information entangled by polyphonic audio events (AEs). However, these approaches have difficulties in explaining what cues they use to identify scenes. This paper conducts the first study on disclosing the relationship between real-life acoustic scenes and semantic embeddings from the most relevant AEs. Specifically, we propose an event-relational graph representation learning (ERGL) framework for ASC to classify scenes, and simultaneously answer clearly and straightly which cues are used in classifying. In the event-relational graph, embeddings of each event are treated as nodes, while relationship cues derived from each pair of nodes are described by multi-dimensional edge features. Experiments on a real-life ASC dataset show that the proposed ERGL achieves competitive performance on ASC by learning embeddings of only a limited number of AEs. The results show the feasibility of recognizing diverse acoustic scenes based on the audio event-relational graph. Visualizations of graph representations learned by ERGL are available here (https://github.com/Yuanbo2020/ERGL).