A statistical model-based voice activity detection

Jongseo Sohn

doi:10.1109/97.736233

A statistical model-based voice activity detection

1999-01-01

1.1K

PRE

AI

DOI：10.1109/97.736233

原文链接

原文求助

摘要

En 中文

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.

Keyword:

decision-directed estimation

hidden Markov model

likelihood ratio test

voice activity detection

A statistical model-based voice activity detection

摘要

期刊

机构

引用论文