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

2025-05-25

APSIPA Transactions on Signal and Information Processing

AI Summary

In industrial acoustic anomaly detection (ASD), limited real anomalous samples severely hinder model generalization. To address this, we propose Serial-Outlier Exposure (Serial-OE), the first framework to adapt Outlier Exposure to ASD: it dynamically incorporates a minimal number (1–5) of authentic anomalies while jointly training on normal data and synthetically generated pseudo-anomalies. Our method employs a deep autoencoder with serial feature modeling, integrating pseudo-anomaly generation, multi-stage loss optimization, and machine-ID-agnostic training. Evaluated on DCASE2020 Task 2, Serial-OE achieves state-of-the-art performance across all metrics. It significantly improves AUC under few-shot settings and demonstrates robustness against label contamination and device-unlabeled data—key challenges in real-world industrial deployment.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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.

Problem

• Detects anomalous sounds using limited anomalous training data
• Improves performance by combining normal and pseudo-anomalous data
• Addresses scarcity of real anomalous data in conventional methods

Innovation

• Serial-OE uses small anomalous data for training
• Outlier exposure with normal and pseudo-anomalous data
• Dynamic adaptation with anomalous data in operation