A double-talk detector is used to freeze acoustic echo canceller's (AEC) filter adaptation during periods of near-end speech. Increased sensitivity towards double-talk results in declaring echo-path changes as double-talk which adversely effects the performance of an AEC as we freeze adaptation when we really need to adapt. Thus, we need an efficient and simple echo-path change detector so as to differentiate any echo-path variations from double-talk condition. In this paper, we derive a novel test statistic for echo-path change detection. The proposed decision statistic detects any echo-path variations, is normalized properly and is computationally very efficient as compared to existing techniques. Simulations demonstrate the efficiency of the proposed algorithm.
M. A. Izbal and S. L. Grant, "A Novel Normalized Cross-Correlation Based Echo-path Change Detector," Proceedings of the IEEE Region 5 Technical Conference, 2007, Institute of Electrical and Electronics Engineers (IEEE), Apr 2007.
The definitive version is available at http://dx.doi.org/10.1109/TPSD.2007.4380390
IEEE Region 5 Technical Conference, 2007
Electrical and Computer Engineering
Keywords and Phrases
Acoustic Echo Canceller; Adaptive Filters; Double-Talk Detector; Echo Suppression; Freeze Adaptation; Near-End Speech
Article - Conference proceedings
© 2007 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.