Acoustic event detection utilizing event class and localization information

Xianjun Xia

Research output: ThesisDoctoral Thesis

143 Downloads (Pure)

Abstract

Acoustic event detection aims to detect the event class and to localize the start and the end times of the audio events in various real world scenarios. The audio events are either monophonic or polyphonic depending on whether they are isolated or overlapping with each other. This thesis proposes deep learning solutions utilizing both the event class and event localization information for improving the detection performance when dealing with monophonic or polyphonic sound events. A data augmentation solution is also proposed that addresses limitations in the available training data when deploying deep learning for acoustic event detection.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • The University of Western Australia
Thesis sponsors
Award date1 Jul 2019
DOIs
Publication statusUnpublished - 2019

Fingerprint

Acoustics
Acoustic waves
Deep learning

Cite this

@phdthesis{556803d4184f403fac13a8f433a41247,
title = "Acoustic event detection utilizing event class and localization information",
abstract = "Acoustic event detection aims to detect the event class and to localize the start and the end times of the audio events in various real world scenarios. The audio events are either monophonic or polyphonic depending on whether they are isolated or overlapping with each other. This thesis proposes deep learning solutions utilizing both the event class and event localization information for improving the detection performance when dealing with monophonic or polyphonic sound events. A data augmentation solution is also proposed that addresses limitations in the available training data when deploying deep learning for acoustic event detection.",
keywords = "acoustic event detection, deep learning, signal processing",
author = "Xianjun Xia",
year = "2019",
doi = "10.26182/5d2d1a2f28223",
language = "English",
school = "The University of Western Australia",

}

Xia, X 2019, 'Acoustic event detection utilizing event class and localization information', Doctor of Philosophy, The University of Western Australia. https://doi.org/10.26182/5d2d1a2f28223

Acoustic event detection utilizing event class and localization information. / Xia, Xianjun.

2019.

Research output: ThesisDoctoral Thesis

TY - THES

T1 - Acoustic event detection utilizing event class and localization information

AU - Xia, Xianjun

PY - 2019

Y1 - 2019

N2 - Acoustic event detection aims to detect the event class and to localize the start and the end times of the audio events in various real world scenarios. The audio events are either monophonic or polyphonic depending on whether they are isolated or overlapping with each other. This thesis proposes deep learning solutions utilizing both the event class and event localization information for improving the detection performance when dealing with monophonic or polyphonic sound events. A data augmentation solution is also proposed that addresses limitations in the available training data when deploying deep learning for acoustic event detection.

AB - Acoustic event detection aims to detect the event class and to localize the start and the end times of the audio events in various real world scenarios. The audio events are either monophonic or polyphonic depending on whether they are isolated or overlapping with each other. This thesis proposes deep learning solutions utilizing both the event class and event localization information for improving the detection performance when dealing with monophonic or polyphonic sound events. A data augmentation solution is also proposed that addresses limitations in the available training data when deploying deep learning for acoustic event detection.

KW - acoustic event detection

KW - deep learning

KW - signal processing

U2 - 10.26182/5d2d1a2f28223

DO - 10.26182/5d2d1a2f28223

M3 - Doctoral Thesis

ER -