A statistical model-based voice activity detection

Author

Sohn, Jongseo ; Kim, Nam Soo ; Sung, Wonyong

Author_Institution

School. of Electr. Eng., Seoul Nat. Univ., South Korea

Volume

6

Issue

1

fYear

1999

Firstpage

1

Lastpage

3

Abstract

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.

Keywords

hidden Markov models; parameter estimation; speech coding; variable rate codes; HMM; SNR; VAD; decision-directed parameter estimation method; first-order Markov process modeling; hang-over scheme; likelihood ratio test; low signal-to-noise ratio; robust voice activity detector; simulation results; speech occurrences; statistical model-based; variable-rate speech coding; vehicular noise environment; voice activity detection; Discrete Fourier transforms; Frequency estimation; Hidden Markov models; Light rail systems; Maximum likelihood estimation; Parameter estimation; Signal to noise ratio; Speech enhancement; Statistics; Testing;

fLanguage

English

Journal_Title

Signal Processing Letters, IEEE

Publisher

ieee

ISSN

1070-9908

Type

jour

DOI

10.1109/97.736233

Filename

736233