audio signal processing

Towards Event-Robust Acoustic Scene Classification

arXiv:2606.06921

summary

The paper presents the Event-Shifted Acoustic Scene (ESAS) dataset, a benchmark that adds foreground sound events to background scenes to test how well acoustic scene classification systems handle unexpected audio events, and shows that current models struggle with this challenge.

Abstract

This paper introduces the Event-Shifted Acoustic Scene (ESAS) dataset, a novel benchmark for evaluating the robustness of Acoustic Scene Classification (ASC) systems against unknown sound events. Existing ASC datasets typically contain recordings of clean and consistent audio, while real-world environments often include diverse and unexpected sound events. To bridge this gap, ESAS simulates real-world acoustic variability by injecting foreground sound events into background scenes with the assistance of large language models. In this work, we present the construction methodology, dataset statistics, and evaluation protocols. Furthermore, a comprehensive evaluation of state-of-the-art ASC systems is conducted using the ESAS benchmark. Experimental results reveal that existing ASC models suffer significant performance degradation when facing the event-shift challenge. The introduction of the ESAS dataset aims to drive future research toward event-robust ASC.

Accepted to Interspeech 2026. The ESAS dataset is available at: https://doi.org/10.5281/zenodo.21317541

Topics & keywords

#acoustic scene classification#robustness#event shift#benchmark dataset#sound events#machine listeningESAS datasetevent-shift challengeforeground event injectionlarge language modelsevaluation protocolsstate-of-the-art ASC models
Towards Event-Robust Acoustic Scene Classification · wovepaper