Anomaly Detection in Raw Audio Using Deep Autoregressive Networks

Files in This Item:
File Description SizeFormat 
insight_publication.pdf239.91 kBAdobe PDFDownload
Title: Anomaly Detection in Raw Audio Using Deep Autoregressive Networks
Authors: Rushe, EllenMacNamee, Brian
Permanent link:
Date: 17-May-2019
Online since: 2020-05-06T08:33:44Z
Abstract: Anomaly detection involves the recognition of patterns outside of what is considered normal, given a certain set of input data. This presents a unique set of challenges for machine learning, particularly if we assume a semi-supervised scenario in which anomalous patterns are unavailable at training time meaning algorithms must rely on non-anomalous data alone. Anomaly detection in time series adds an additional level of complexity given the contextual nature of anomalies. For time series modelling, autoregressive deep learning architectures such as WaveNet have proven to be powerful generative models, specifically in the field of speech synthesis. In this paper, we propose to extend the use of this type of architecture to anomaly detection in raw audio. In experiments using multiple audio datasets we compare the performance of this approach to a baseline autoencoder model and show superior performance in almost all cases.
Funding Details: Science Foundation Ireland
metadata.dc.description.othersponsorship: Insight Research Centre
Type of material: Conference Publication
Publisher: IEEE
Copyright (published version): 2019 IEEE
Keywords: Anomaly detectionDeep learningRaw audioWaveNet
DOI: 10.1109/ICASSP.2019.8683414
Language: en
Status of Item: Peer reviewed
Is part of: 2019 IEEE International Conference on Acoustics, Speech and Signal Processing: Proceedings
Conference Details: The 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, Brighton, United Kingdom, 12-17 May 2019
ISBN: 978-1-4799-8131-1
Appears in Collections:Insight Research Collection

Show full item record

Page view(s)

Last Week
Last month
checked on Jun 7, 2020


checked on Jun 7, 2020

Google ScholarTM



This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.