Voice activity detection using entropy in spectrum domain

Meysam Asgari; Abolghasem Sayadian; Mohsen Farhadloo; Elahe Abouie Mehrizi

doi:10.1109/ATNAC.2008.4783359

Voice activity detection using entropy in spectrum domain

Meysam Asgari, Abolghasem Sayadian, Mohsen Farhadloo, Elahe Abouie Mehrizi

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

11 Scopus citations

Abstract

In this paper we develop a voice activity detection algorithm based on entropy estimation of magnitude spectrum. In addition, the likelihood ratio test (LRT) is employed to determine a threshold to separate of speech segments from Nonspeech segments. The distributions of entropy magnitude of clean speech and noise signal are assumed to be Gaussian. The application of the concept of entropy to the speech detection problem is based on the assumption that the signal spectrum is more organized during speech segments than during noise segments. One of the main advantages of this method is that it is not very sensitive to the changes of noise level. Our simulation results show that the entropy based VAD is high performance in low Signal to Noise Ratio (SNR) conditions (SNR < 0 dB).

Original language	English (US)
Title of host publication	Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008
Pages	407-410
Number of pages	4
DOIs	https://doi.org/10.1109/ATNAC.2008.4783359
State	Published - 2008
Externally published	Yes
Event	2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008 - Adelaide, SA, Australia Duration: Dec 7 2008 → Dec 10 2008

Publication series

Name	Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008

Conference

Conference	2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008
Country/Territory	Australia
City	Adelaide, SA
Period	12/7/08 → 12/10/08

ASJC Scopus subject areas

Computer Networks and Communications
Electrical and Electronic Engineering
Communication

Access to Document

10.1109/ATNAC.2008.4783359

Cite this

Asgari, M., Sayadian, A., Farhadloo, M., & Mehrizi, E. A. (2008). Voice activity detection using entropy in spectrum domain. In Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008 (pp. 407-410). Article 4783359 (Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008). https://doi.org/10.1109/ATNAC.2008.4783359

Voice activity detection using entropy in spectrum domain. / Asgari, Meysam; Sayadian, Abolghasem; Farhadloo, Mohsen et al.
Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008. 2008. p. 407-410 4783359 (Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008).

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

Asgari, M, Sayadian, A, Farhadloo, M & Mehrizi, EA 2008, Voice activity detection using entropy in spectrum domain. in Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008., 4783359, Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008, pp. 407-410, 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008, Adelaide, SA, Australia, 12/7/08. https://doi.org/10.1109/ATNAC.2008.4783359

Asgari M, Sayadian A, Farhadloo M, Mehrizi EA. Voice activity detection using entropy in spectrum domain. In Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008. 2008. p. 407-410. 4783359. (Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008). doi: 10.1109/ATNAC.2008.4783359

@inproceedings{4618a635d7bb49d787f43657785490a3,

title = "Voice activity detection using entropy in spectrum domain",

abstract = "In this paper we develop a voice activity detection algorithm based on entropy estimation of magnitude spectrum. In addition, the likelihood ratio test (LRT) is employed to determine a threshold to separate of speech segments from Nonspeech segments. The distributions of entropy magnitude of clean speech and noise signal are assumed to be Gaussian. The application of the concept of entropy to the speech detection problem is based on the assumption that the signal spectrum is more organized during speech segments than during noise segments. One of the main advantages of this method is that it is not very sensitive to the changes of noise level. Our simulation results show that the entropy based VAD is high performance in low Signal to Noise Ratio (SNR) conditions (SNR < 0 dB).",

author = "Meysam Asgari and Abolghasem Sayadian and Mohsen Farhadloo and Mehrizi, {Elahe Abouie}",

note = "Copyright: Copyright 2009 Elsevier B.V., All rights reserved.; 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008 ; Conference date: 07-12-2008 Through 10-12-2008",

year = "2008",

doi = "10.1109/ATNAC.2008.4783359",

language = "English (US)",

isbn = "9781424426034",

series = "Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008",

pages = "407--410",

booktitle = "Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008",

}

TY - GEN

T1 - Voice activity detection using entropy in spectrum domain

AU - Asgari, Meysam

AU - Sayadian, Abolghasem

AU - Farhadloo, Mohsen

AU - Mehrizi, Elahe Abouie

PY - 2008

Y1 - 2008

N2 - In this paper we develop a voice activity detection algorithm based on entropy estimation of magnitude spectrum. In addition, the likelihood ratio test (LRT) is employed to determine a threshold to separate of speech segments from Nonspeech segments. The distributions of entropy magnitude of clean speech and noise signal are assumed to be Gaussian. The application of the concept of entropy to the speech detection problem is based on the assumption that the signal spectrum is more organized during speech segments than during noise segments. One of the main advantages of this method is that it is not very sensitive to the changes of noise level. Our simulation results show that the entropy based VAD is high performance in low Signal to Noise Ratio (SNR) conditions (SNR < 0 dB).

AB - In this paper we develop a voice activity detection algorithm based on entropy estimation of magnitude spectrum. In addition, the likelihood ratio test (LRT) is employed to determine a threshold to separate of speech segments from Nonspeech segments. The distributions of entropy magnitude of clean speech and noise signal are assumed to be Gaussian. The application of the concept of entropy to the speech detection problem is based on the assumption that the signal spectrum is more organized during speech segments than during noise segments. One of the main advantages of this method is that it is not very sensitive to the changes of noise level. Our simulation results show that the entropy based VAD is high performance in low Signal to Noise Ratio (SNR) conditions (SNR < 0 dB).

UR - http://www.scopus.com/inward/record.url?scp=64049109501&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=64049109501&partnerID=8YFLogxK

U2 - 10.1109/ATNAC.2008.4783359

DO - 10.1109/ATNAC.2008.4783359

M3 - Conference contribution

AN - SCOPUS:64049109501

SN - 9781424426034

T3 - Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008

SP - 407

EP - 410

BT - Proceedings of the 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008

T2 - 2008 Australasian Telecommunication Networks and Applications Conference, ATNAC 2008

Y2 - 7 December 2008 through 10 December 2008

ER -

Voice activity detection using entropy in spectrum domain

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this