Multi-Microphone acoustic echo cancellation using relative echo transfer functions

摘要

Modern hands-free communication devices, such as smart speakers, are equipped with several microphones, and one or more loudspeakers. The most straightforward solution to reduce acoustic echoes is to apply acoustic echo cancellation (AEC) to each microphone. Due to limited computational resources, the implementation of such a solution may not be feasible. To overcome this problem, a method is proposed that uses a primary estimated echo signal, obtained using state-of-the-art AEC, to compute the remaining, or secondary, acoustic echoes. To do this, relative transfer functions between primary and secondary acoustic echo signals, referred to as relative echo transfer functions (RETFs), are estimated and employed. In this work, the acoustic echo transfer functions (AETFs) and RETFs are modeled using convolutive transfer functions. Provided that the distance between microphones is small, the RETFs can be modeled using fewer partitions than the AETFs, which reduces the overall computational complexity.