A survey of double-talk detection schemes for echo cancellation applications

Authors

  • Thien-An Vu Sch. of Info. Technol. and Eng., University of Ottawa, Ont., Canada
  • Heping Ding Acoustics and Signal Processing, IMS, National Research Council, Ottawa, Ont., Canada
  • Martin Bouchard Sch. of Info. Technol. and Eng., University of Ottawa, Ont., Canada

Keywords:

Adaptive filtering, Algorithms, Computational complexity, Echo suppression, Impulse response, Matrix algebra, Parameter estimation, Signal detection, Signal receivers, Vectors, Double-talk detection, Echo paths, Energy-based algorithms, Filter adaptation

Abstract

Double-talk (DT) detection schemes for echo cancellation based on energy comparison and cross-correlation algorithms are presented. It was found that the correlation-based algorithms show improved detection performance in such time-varying environment. These algorithms however require high memory storage and computational complexity due to vector or matrix-based operations. The energy-based algorithms are computationally simple, needing very little memory, but they do not always provide reliable performance in an acoustic echo path environment.

Downloads

Published

2004-09-01

How to Cite

1.
Vu T-A, Ding H, Bouchard M. A survey of double-talk detection schemes for echo cancellation applications. Canadian Acoustics [Internet]. 2004 Sep. 1 [cited 2021 Dec. 8];32(3):144-5. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/1669

Issue

Section

Proceedings of the Acoustics Week in Canada