Impact of Acoustic Event Tagging on Scene Classification in a Multi-Task Learning Framework

TL;DR

This study investigates how acoustic event tagging influences scene classification within a multi-task learning framework, revealing that AET can enhance ASC performance through regularization effects rather than direct event discrimination.

Contribution

The paper provides an extensive empirical analysis demonstrating conditions under which AET improves ASC, emphasizing the regularization role of AET in multi-task networks.

Findings

AET as an auxiliary task can improve ASC performance.

ASC gains increase with AET dataset size.

Performance is insensitive to specific event choices in AET.

Abstract

Acoustic events are sounds with well-defined spectro-temporal characteristics which can be associated with the physical objects generating them. Acoustic scenes are collections of such acoustic events in no specific temporal order. Given this natural linkage between events and scenes, a common belief is that the ability to classify events must help in the classification of scenes. This has led to several efforts attempting to do well on Acoustic Event Tagging (AET) and Acoustic Scene Classification (ASC) using a multi-task network. However, in these efforts, improvement in one task does not guarantee an improvement in the other, suggesting a tension between ASC and AET. It is unclear if improvements in AET translates to improvements in ASC. We explore this conundrum through an extensive empirical study and show that under certain conditions, using AET as an auxiliary task in the multi-task network consistently improves ASC performance. Additionally, ASC performance further improves with the AET data-set size and is not sensitive to the choice of events or the number of events in the AET data-set. We conclude that this improvement in ASC performance comes from the regularization effect of using AET and not from the network's improved ability to discern between acoustic events.