Signal Subspace-based Voice Activity Detection Using Generalized Gaussian Distribution

Yong-Sub Um; Joon-Hyuk Chang; Dong Kook Kim

doi:10.7776/ASK.2013.32.2.131

All Issue

2013 Vol.32, Issue 2 Preview Page Next Page

Signal Subspace-based Voice Activity Detection Using Generalized Gaussian Distribution 일반화된 가우시안 분포를 이용한 신호 준공간 기반의 음성검출기법

31 March 2013. pp. 131-137

PDF XML

Abstract

In this paper we propose an improved voice activity detection (VAD) algorithm using statistical models in the signal subspace domain. A uncorrelated signal subspace is generated using embedded prewhitening technique and the statistical characteristics of the noisy speech and noise are investigated in this domain. According to the characteristics of the signals in the signal subspace, a new statistical VAD method using GGD (Generalized Gaussian Distribution) is proposed. Experimental results show that the proposed GGD-based approach outperforms the Gaussian-based signal subspace method at 0-15 dB SNR simulation conditions.

Keywords

Voice activity detection

Signal subspace

Generalized gaussian distribution

본 논문에서는 신호준공간 (signal subspace) 영역에서 통계적 모델을 이용한 향상된 음성검출기법을 제안한다. 이를 위해 EP(Embedded Prewhitening) 기법에 의해 비상관적인 (uncorrelated) 신호준공간을 생성하고, 이 영역에서 잡음음성과 잡음에 대한 통계적 특성을 파악하였다. 이러한 통계적 특성에 근거하여 GGD (Generalized Gaussian Distribution)을 사용하여 보다 효율적인 음성검출 알고리즘을 제안한다. 실험을 통해 제안된 기법이 0-15dB SNR의 시뮬레이션 환경에서 기존 Gaussian을 사용한 신호준공간 기법보다 향상된 음성검출 결과를 보여준다.

키워드

음성검출

신호 준공간

일반적인 가우시안 분포

References

J. S. Sohn, N. S. Kim, and W. Y. Sung, "A statistical model-based voice activity detection," IEEE Sig. Proc. Lett. 6, 1-3 (1999).

10.1109/97.736233

Y. D. Cho and A. Kondoz, "Analysis and improve-ment of a statistical model-based voice activity de-tector," IEEE Sig. Proc. Lett. 8, 276-278 (2001).

10.1109/97.957270

J. H. Chang, N. S. Kim, and S. K. Mitra, "Voice activity detection based on multiple statistical models," IEEE Trans. Signal Proc. 54, 1965-1976 (2006).

10.1109/TSP.2006.874403

J. H. Chang, J. W. Shin, and N. S. Kim, "Voice activity detector employing generalized Gaussian dis-tribution," IEE Electronics Lett. 40, 1561-1563 (2004).

10.1049/el:20047090

J. W. Shin, J. H. Chang, and N. S. Kim, "Statistical modeling of speech signals based on generalized Gamma distribution," IEEE Sig. Proc. Lett. 12, 258- 261 (2005).

10.1109/LSP.2004.840869

S. Gazor and W. Zhang, "A Soft Voice Activity Detector Based on a Laplacian-Gaussian Model," IEEE Trans. Speech and Audio Proc. 11, 498-505 (2003).

10.1109/TSA.2003.815518

K. C. Ryu and D. K. Kim, "Statistical voice activity detector based on signal subspace model," (In Korean), J. Acoust. Soc. Kr. 27, 372-378 (2008).

D. K. Kim and J.H. Chang, "A subspace approach based on embedded prewhitening for voice activity detection," J. Acoust. Soc. Am. 130, 304-310 (2011).

10.1121/1.363892722088032

Y. S. Um, J. H. Chang, and D. K. Kim "An improved VAD approach based on generalized Gaussian distribution in signal subspace domain," (In Korean), J. Acoust. Soc. Kr. 2(s) 30, 136-139 (2011).

S. Gazor and W. Zhang "Speech probability dis--tribution," IEEE Sig. Proc. Lett. 10, 204-207 (2003).

10.1109/LSP.2003.813679

Y. Hu and P. C. Loizou, "A generalized subspace approach for enhancing speech corrupted by colored noise," IEEE Trans. Speech Audio Proc. 11, 334-341 (2003).

10.1109/TSA.2003.814458

A. G. Glen, L. M. Leemis, and D. R. Barr, "Order statistics in goodness-of-fit testing," IEEE Trans. Reliab. 50, 209-213 (2001).

10.1109/24.963129

S. Nadarajah, "A generalized Gaussian distribution," J. Appl. Stat. 32, 685- 694 (2005).

10.1080/02664760500079464

Information

Publisher :The Acoustical Society of Korea
Publisher(Ko) :한국음향학회
Journal Title :The Journal of the Acoustical Society of Korea
Journal Title(Ko) :한국음향학회지
Volume : 32
No :2
Pages :131-137
Received Date : 2012-09-14
Revised Date : 2012-10-19
Accepted Date : 2012-10-19
DOI :https://doi.org/10.7776/ASK.2013.32.2.131

The Journal of the Acoustical Society of KoreaISSN:1225-4428(Print) 2287-3775(Online)한국음향학회

All Issue