Description and Discussion on DCASE 2026 Challenge Task 2: Noise-aware Unsupervised Anomalous Sound Detection for Machine Condition Monitoring

2026-05-31

AI Summary

This work addresses the challenge of unsupervised anomalous sound detection (UASD) under strong environmental noise using only normal samples. To enhance robustness in realistic noisy conditions, the authors propose a novel approach leveraging dual-channel audio inputs—signals from both near-end and far-end microphones. By incorporating the far-end microphone signal, the method improves perception of ambient noise, enabling more effective separation of background interference from target machine sounds. Notably, this study introduces the first dual-channel recording setup in the DCASE challenge, establishing a new data paradigm and a noise-aware modeling strategy for UASD. Experimental results demonstrate that the proposed method significantly improves detection performance, validating its effectiveness in real-world noisy environments.

Abstract

This paper presents an overview of DCASE 2026 Challenge Task 2, titled "Noise-aware unsupervised anomalous sound detection (UASD) for machine condition monitoring." The task aims to advance noise-robust anomalous sound detection for machine condition monitoring under the unsupervised setting, where only normal machine sounds are available for training. Reliable detection under noisy conditions is crucial for practical deployment, but previous DCASE Task 2 settings provided limited information about environmental noise, potentially limiting UASD performance in highly noisy situations. To address this limitation, DCASE 2026 allows participants to exploit two-channel audio samples simultaneously captured at locations near and far from the target machine. Since the distant microphone is expected to contain relatively stronger environmental noise and weaker direct machine sounds, it may help distinguish environmental noise components from the target machine sounds. After the challenge submission deadline, challenge results and an analysis of the submitted systems will be added.

Problem

• anomalous sound detection
• machine condition monitoring
• noise-aware
• unsupervised learning
• environmental noise

Innovation

• noise-aware
• unsupervised anomalous sound detection
• two-channel audio
• machine condition monitoring
• environmental noise separation