A statistical model-based voice activity detection

被引：932

作者：

Sohn, J ^{[1
]}

Kim, NS ^{[1
]}

Sung, W ^{[1
]}

机构：

[1] Seoul Natl Univ, Sch Elect Engn, Seoul 151742, South Korea

来源：

IEEE SIGNAL PROCESSING LETTERS | 1999年 / 6卷 / 01期

关键词：

decision-directed estimation; hidden Markov model; likelihood ratio test; voice activity detection;

D O I：

10.1109/97.736233

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

In this letter, we develop a robust voice activity detector (VAD) for the application to variable-rate speech coding. The developed VAD employs the decision-directed parameter estimation method for the likelihood ratio test. In addition, we propose an effective hang-over scheme which considers the previous observations by a first-order Markov process modeling of speech occurrences. According to our simulation results, the proposed VAD shows significantly better performances than the G.729B VAD in low signal-to-noise ratio (SNR) and vehicular noise environments.

引用

页码：1 / 3

页数：3

共 5 条

[1] Elimination of the Musical Noise Phenomenon with the Ephraim and Malah Noise Suppressor [J].

Cappe, Olivier .

IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 1994, 2 (02) :345-349

[2] SPEECH ENHANCEMENT USING A MINIMUM MEAN-SQUARE ERROR SHORT-TIME SPECTRAL AMPLITUDE ESTIMATOR [J].

EPHRAIM, Y ;

MALAH, D .

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1984, 32 (06) :1109-1121

[3]

Rabiner L., 1993, Fundamentals of Speech Recognition

[4]

Sohn J, 1998, INT CONF ACOUST SPEE, P365, DOI 10.1109/ICASSP.1998.674443

[5]

Srinivasan K., 1993, IEEE SPEECH COD WORK

← 1 →