Skip to main content

Robust Sound Event Detection in Continuous Audio Environments

Zhang, Hao-min and McLoughlin, Ian Vince and Song, Yan (2016) Robust Sound Event Detection in Continuous Audio Environments. In: 17th Annual Conference of the International Speech Communication Association (INTERSPEECH 2016): Understanding Speech Processing in Humans and Machines. Curran Associates, Red Hook, New York, USA. ISBN 978-1-5108-3313-5. (doi:10.21437/Interspeech.2016-392)

Abstract

Sound event detection in real world environments has attracted significant research interest recently because of it's applications in popular fields such as machine hearing and automated surveillance, as well as in sound scene understanding. This paper considers continuous robust sound event detection, which means multiple overlapped sound events in different types of interfering noise. First, a standard evaluation task is outlined based upon existing testing data sets for the sound event classification of isolated sounds. This paper then proposes and evaluates the use of spectrogram image features employing an energy detector to segment sound events, before developing a novel segmentation method making use of a Bayesian inference criteria. At the back end, a convolutional neural network is used to classify detected regions, and this combination is compared to several alternative approaches. The proposed method is shown capable of achieving very good performance compared with current state-of-the-art techniques.

Item Type: Book section
DOI/Identification number: 10.21437/Interspeech.2016-392
Uncontrolled keywords: sound event detection, convolutional neural network, Bayesian inference, segmentation
Subjects: T Technology
Divisions: Faculties > Sciences > School of Computing > Data Science
Depositing User: Ian McLoughlin
Date Deposited: 15 Jul 2016 08:24 UTC
Last Modified: 26 Sep 2019 11:14 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/56309 (The current URI for this page, for reference purposes)
McLoughlin, Ian Vince: https://orcid.org/0000-0001-7111-2008
  • Depositors only (login required):

Downloads

Downloads per month over past year