APSIPA Transactions on Signal and Information Processing > Vol 14 > Issue 1

Serial-OE: Anomalous Sound Detection Based on Serial Method with Outlier Exposure Capable of Using Small Amounts of Anomalous Data for Training

Ibuki Kuroyanagi, Nagoya University, Japan, kuroyanagi.ibuki@g.sp.m.is.nagoya-u.ac.jp , Tomoki Hayashi, Nagoya University, Japan, Kazuya Takeda, Nagoya University, Japan, Tomoki Toda, Nagoya University, Japan
 
Suggested Citation
Ibuki Kuroyanagi, Tomoki Hayashi, Kazuya Takeda and Tomoki Toda (2025), "Serial-OE: Anomalous Sound Detection Based on Serial Method with Outlier Exposure Capable of Using Small Amounts of Anomalous Data for Training", APSIPA Transactions on Signal and Information Processing: Vol. 14: No. 1, e1. http://dx.doi.org/10.1561/116.20240060

Publication Date: 21 Jan 2025
© 2025 I. Kuroyanagi, T. Hayashi, K. Takeda and T. Toda
 
Subjects
Detection and estimation,  Pattern recognition and learning,  Feature detection and selection,  Audio signal processing,  Classification and prediction,  Deep learning
 
Keywords
Anomalous sound detectionself-supervised learningoutlier exposureserial method
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 68 times

In this article:
Introduction 
ASD Methods that can Utilize Anomalous Data 
Proposed Method 
Experimental Evaluations 
Limitations 
Conclusion 
References 

Abstract

We introduce Serial-OE, a new approach to anomalous sound detection (ASD) that leverages small amounts of anomalous data to improve the performance. Conventional ASD methods rely primarily on the modeling of normal data, due to the cost of collecting anomalous data from various possible types of equipment breakdowns. Our method improves upon existing ASD systems by implementing an outlier exposure framework that utilizes normal and pseudo-anomalous data for training, with the capability to also use small amounts of real anomalous data. A comprehensive evaluation using the DCASE2020 Task2 dataset shows that our method outperforms state-of-the-art ASD models. We also investigate the impact on performance of using a small amount of anomalous data during training, of using data without machine ID information, and of using contaminated training data. Our experimental results reveal the potential of using a very limited amount of anomalous data during training to address the limitations of existing methods using only normal data for training due to the scarcity of anomalous data. This study contributes to the field by presenting a method that can be dynamically adapted to include anomalous data during the operational phase of an ASD system, paving the way for more accurate ASD.

DOI:10.1561/116.20240060