This paper proposes a novel optimization principle and its implementation for unsupervised anomaly detection in sound (ADS) using an autoencoder (AE). The goal of unsupervised-ADS is to detect unknown anomalous sound without training data of anomalous sound. Use of an AE as a normal model is a state-of-the-art technique for unsupervised-ADS. To decrease the false positive rate (FPR), the AE is trained to minimize the reconstruction error of normal sounds and the anomaly score is calculated as the reconstruction error of the observed sound. Unfortunately, since this training procedure does not take into account the anomaly score for anomalous sounds, the true positive rate (TPR) does not necessarily increase. In this study, we define an objective function based on the Neyman-Pearson lemma by considering ADS as a statistical hypothesis test. The proposed objective function trains the AE to maximize the TPR under an arbitrary low FPR condition. To calculate the TPR in the objective function, we consider that the set of anomalous sounds is the complementary set of normal sounds and simulate anomalous sounds by using a rejection sampling algorithm. Through experiments using synthetic data, we found that the proposed method improved the performance measures of ADS under low FPR conditions. In addition, we confirmed that the proposed method could detect anomalous sounds in real environments.Index Terms-Anomaly detection in sound, Neyman-Pearson lemma, deep learning, and autoencoder.All authors are with the ). A preliminary version of this work is published in [8].

This paper introduces a new dataset called "ToyADMOS" designed for anomaly detection in machine operating sounds (ADMOS). To the best our knowledge, no large-scale datasets are available for ADMOS, although large-scale datasets have contributed to recent advancements in acoustic signal processing. This is because anomalous sound data are difficult to collect. To build a large-scale dataset for ADMOS, we collected anomalous operating sounds of miniature machines (toys) by deliberately damaging them. The released dataset consists of three sub-datasets for machine-condition inspection, fault diagnosis of machines with geometrically fixed tasks, and fault diagnosis of machines with moving tasks. Each sub-dataset includes over 180 hours of normal machine-operating sounds and over 4,000 samples of anomalous sounds collected with four microphones at a 48-kHz sampling rate. The dataset is freely available for download at https://github.com/YumaKoizumi/ ToyADMOS-dataset.Index Terms-Anomaly detection in sounds, machine operating sounds, product inspection, dataset.

Abstract-We propose a method for optimizing an acoustic feature extractor for anomalous sound detection (ASD). Most ASD systems adopt outlier-detection techniques because it is difficult to collect a massive amount of anomalous sound data. To improve the performance of such outlier-detection-based ASD, it is essential to extract a set of efficient acoustic features that is suitable for identifying anomalous sounds. However, the ideal property of a set of acoustic features that maximizes ASD performance has not been clarified. By considering outlierdetection-based ASD as a statistical hypothesis test, we defined optimality as an objective function that adopts Neyman-Pearson lemma; the acoustic feature extractor is optimized to extract a set of acoustic features which maximize the true positive rate under an arbitrary false positive rate. The variational auto-encoder is applied as an acoustic feature extractor and optimized to maximize the objective function. We confirmed that the proposed method improved the F-measure score from 0.02 to 0.06 points compared to those of conventional methods, and ASD results of a stereolithography 3D-printer in a real-environment show that the proposed method is effective in identifying anomalous sounds.

No abstract

No abstract