Anomalous Sound Detection with Machine Learning: A Systematic Review

TL;DR

This systematic review analyzes machine learning techniques for anomalous sound detection, highlighting datasets, feature extraction methods, models, and evaluation metrics used in recent studies from 2010 to 2020.

Contribution

It provides a comprehensive overview of the state-of-the-art ML approaches, datasets, and evaluation methods for anomalous sound detection in a systematic manner.

Findings

ToyADMOS, MIMII, and Mivia datasets are most used.

MFCC is the preferred feature extraction method.

Autoencoder and CNN are the most cited ML models.

Abstract

Anomalous sound detection (ASD) is the task of identifying whether the sound emitted from an object is normal or anomalous. In some cases, early detection of this anomaly can prevent several problems. This article presents a Systematic Review (SR) about studies related to Anamolous Sound Detection using Machine Learning (ML) techniques. This SR was conducted through a selection of 31 (accepted studies) studies published in journals and conferences between 2010 and 2020. The state of the art was addressed, collecting data sets, methods for extracting features in audio, ML models, and evaluation methods used for ASD. The results showed that the ToyADMOS, MIMII, and Mivia datasets, the Mel-frequency cepstral coefficients (MFCC) method for extracting features, the Autoencoder (AE) and Convolutional Neural Network (CNN) models of ML, the AUC and F1-score evaluation methods were most cited.