Few-Shot Adaptation for Anomalous Sound Detection at Low False-Positive Rate
Authors: Simone Lugani, Christof Pichler, Hannes Wegleiter, Christoph Feilmayr, Rodolfo Zunino, Edoardo Ragusa
Organizations: DITEN, University of Genoa, Genoa, Italy · Christian Doppler Laboratory for Measurement Systems for Harsh Operating Conditions, EMS, TU Graz, Graz, Austria · voestalpine Stahl GmbH, Linz, Austria
Reliable fault detection is crucial for industrial machinery in large-scale facilities, where unexpected failures can cause costly downtime and production interruptions. Although acoustic monitoring provides a scalable solution for machine condition monitoring, its deployment is often constrained by high false-positive rates. We propose a two-step approach to automatic anomalous sound detection: first, when only normal operating audio samples are available, a regularized unsupervised model is deployed; then, after the first machine fault, a few-shot adaptation strategy exploits the anomalous samples acquired during the fault to improve the system. This adaptation increases the area under the precision-recall curve (AUPR) from 23.0 to 64.0 and improves detection in the low false-positive-rate region, with the partial area under the receiver operating characteristic curve (pAUROC) increasing from 39.6 to 72.8. We also show that covariance regularization improves both prediction accuracy and training stability in the unsupervised model. These results provide a practical strategy for acoustic anomaly detection systems, enabling improved accuracy and reduced false alarms.
Figures & tables
Fig. 1: Block diagram of the ASD framework: the log-mel spectrogram is compressed by the CVAE; then, a GMM is used to compute the anomaly score based on vector similarity with the learned normal representation.
Fig. 2: Overview of the measurement setup and recreated fault case: facility layout with microphone and fault locations, vibrating screening machine, and physical reconstruction of typical mechanical failures.
Fig. 3: AUPR performance of the ASD model with regularization after the unsupervised training step (UNS_R) and after few-shot adaptation (FS), compared with the unsupervised state-of-the-art approach without regularization (UNS). The few-shot adaptation step yields a drastic improvement.
Model
pAUROC
TPR@FPR 1%
TPR@FPR 0.1%
UNS
35.5±1.2
58.0±1.4
7.5±0.8
UNS_R
39.6±0.9
64.4±1.0
9.4±0.8
FS
72.8±0.4
89.3±0.0
60.6±0.1
TABLE I: Performance comparison of the models (mean ± std)
Information Sciences, CAI-3, Los Alamos National Laboratory, Los Alamos, USA · Information Sciences (CAI-3), Los Alamos National Laboratory, Los Alamos, NM, USA · AeroVironment Inc., Albuquerque, USA +7