Isolated word endpoint detection using time-frequency variance kernels

Alexandros Kyriakides; Costas Pitris; Alex Fink; Andreas Spanias

doi:10.1109/ACSSC.2011.6190170

Isolated word endpoint detection using time-frequency variance kernels

Alexandros Kyriakides, Costas Pitris, Alex Fink, Andreas Spanias

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

3 Scopus citations

Abstract

A major challenge in developing endpoint detection systems is the presence of background noise. We have developed a hybrid method for performing endpoint detection which is based on spectrogram estimation using LPC and a detection process based on imaging operations on the spectrogram. High-variance regions in the spectrogram, captured by variance kernels, can be used to accurately determine the endpoints of speech. This hybrid approach to endpoint detection is robust to various types and levels of background noise. Compared with two other publicly-available methods, our approach performs favorably.

Original language	English (US)
Title of host publication	Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011
Pages	1041-1045
Number of pages	5
DOIs	https://doi.org/10.1109/ACSSC.2011.6190170
State	Published - Dec 1 2011
Event	45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011 - Pacific Grove, CA, United States Duration: Nov 6 2011 → Nov 9 2011

Publication series

Name	Conference Record - Asilomar Conference on Signals, Systems and Computers
ISSN (Print)	1058-6393

Other

Other	45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011
Country/Territory	United States
City	Pacific Grove, CA
Period	11/6/11 → 11/9/11

ASJC Scopus subject areas

Signal Processing
Computer Networks and Communications

Access to Document

10.1109/ACSSC.2011.6190170

Cite this

Kyriakides, A., Pitris, C., Fink, A., & Spanias, A. (2011). Isolated word endpoint detection using time-frequency variance kernels. In Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011 (pp. 1041-1045). [6190170] (Conference Record - Asilomar Conference on Signals, Systems and Computers). https://doi.org/10.1109/ACSSC.2011.6190170

Isolated word endpoint detection using time-frequency variance kernels. / Kyriakides, Alexandros; Pitris, Costas; Fink, Alex et al.
Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011. 2011. p. 1041-1045 6190170 (Conference Record - Asilomar Conference on Signals, Systems and Computers).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kyriakides, A, Pitris, C, Fink, A & Spanias, A 2011, Isolated word endpoint detection using time-frequency variance kernels. in Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011., 6190170, Conference Record - Asilomar Conference on Signals, Systems and Computers, pp. 1041-1045, 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011, Pacific Grove, CA, United States, 11/6/11. https://doi.org/10.1109/ACSSC.2011.6190170

@inproceedings{38569ebee18241d0ae6d8e810ffe19ed,

title = "Isolated word endpoint detection using time-frequency variance kernels",

abstract = "A major challenge in developing endpoint detection systems is the presence of background noise. We have developed a hybrid method for performing endpoint detection which is based on spectrogram estimation using LPC and a detection process based on imaging operations on the spectrogram. High-variance regions in the spectrogram, captured by variance kernels, can be used to accurately determine the endpoints of speech. This hybrid approach to endpoint detection is robust to various types and levels of background noise. Compared with two other publicly-available methods, our approach performs favorably.",

author = "Alexandros Kyriakides and Costas Pitris and Alex Fink and Andreas Spanias",

year = "2011",

month = dec,

day = "1",

doi = "10.1109/ACSSC.2011.6190170",

language = "English (US)",

isbn = "9781467303231",

series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",

pages = "1041--1045",

booktitle = "Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011",

note = "45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011 ; Conference date: 06-11-2011 Through 09-11-2011",

}

TY - GEN

T1 - Isolated word endpoint detection using time-frequency variance kernels

AU - Kyriakides, Alexandros

AU - Pitris, Costas

AU - Fink, Alex

AU - Spanias, Andreas

PY - 2011/12/1

Y1 - 2011/12/1

N2 - A major challenge in developing endpoint detection systems is the presence of background noise. We have developed a hybrid method for performing endpoint detection which is based on spectrogram estimation using LPC and a detection process based on imaging operations on the spectrogram. High-variance regions in the spectrogram, captured by variance kernels, can be used to accurately determine the endpoints of speech. This hybrid approach to endpoint detection is robust to various types and levels of background noise. Compared with two other publicly-available methods, our approach performs favorably.

AB - A major challenge in developing endpoint detection systems is the presence of background noise. We have developed a hybrid method for performing endpoint detection which is based on spectrogram estimation using LPC and a detection process based on imaging operations on the spectrogram. High-variance regions in the spectrogram, captured by variance kernels, can be used to accurately determine the endpoints of speech. This hybrid approach to endpoint detection is robust to various types and levels of background noise. Compared with two other publicly-available methods, our approach performs favorably.

UR - http://www.scopus.com/inward/record.url?scp=84861316333&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861316333&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2011.6190170

DO - 10.1109/ACSSC.2011.6190170

M3 - Conference contribution

AN - SCOPUS:84861316333

SN - 9781467303231

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 1041

EP - 1045

BT - Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011

T2 - 45th Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2011

Y2 - 6 November 2011 through 9 November 2011

ER -

Isolated word endpoint detection using time-frequency variance kernels

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this