Random forest classification based acoustic event detection utilizing contextual-information and bottleneck features

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30% and 4.44% absolute error rate improvement respectively compared to the state of art technique.

Original languageEnglish
Pages (from-to)1-13
Number of pages13
JournalPattern Recognition
Volume81
DOIs
Publication statusPublished - 1 Sep 2018

Fingerprint

Acoustics

Cite this

@article{8ddccf72b6c3462cacf739e963526d14,
title = "Random forest classification based acoustic event detection utilizing contextual-information and bottleneck features",
abstract = "The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30{\%} and 4.44{\%} absolute error rate improvement respectively compared to the state of art technique.",
keywords = "Acoustic event detection, Category-specific bottleneck features, Contextual information, Global bottleneck features",
author = "Xianjun Xia and Roberto Togneri and Ferdous Sohel and David Huang",
year = "2018",
month = "9",
day = "1",
doi = "10.1016/j.patcog.2018.03.025",
language = "English",
volume = "81",
pages = "1--13",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier",

}

TY - JOUR

T1 - Random forest classification based acoustic event detection utilizing contextual-information and bottleneck features

AU - Xia, Xianjun

AU - Togneri, Roberto

AU - Sohel, Ferdous

AU - Huang, David

PY - 2018/9/1

Y1 - 2018/9/1

N2 - The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30% and 4.44% absolute error rate improvement respectively compared to the state of art technique.

AB - The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30% and 4.44% absolute error rate improvement respectively compared to the state of art technique.

KW - Acoustic event detection

KW - Category-specific bottleneck features

KW - Contextual information

KW - Global bottleneck features

UR - http://www.scopus.com/inward/record.url?scp=85044653973&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2018.03.025

DO - 10.1016/j.patcog.2018.03.025

M3 - Article

VL - 81

SP - 1

EP - 13

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

ER -