A statistical model-based voice activity detection

Download PDF
Open Webpage

Jongseo Sohn, Nam Soo Kim, Wonyong Sung

Published: 1999, Last Modified: 09 Jan 2026IEEE Signal Process. Lett. 1999EveryoneRevisionsBibTeXCC BY-SA 4.0
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.