The acoustic echo cancelation using blind source separation to reduce double-talk interference

An acoustic echo canceler (AEC) has been used to remove acoustic echoes generated in hands-free telecommunications. The AEC cancels the acoustic echoes by approximating the echo-path with use of an adaptive filter and subtracting the pseudo echoes generated by the filter from the observed signal. However, the conventional adaptive algorithm for updating the filter fails in estimation of the echo-path during double-talk in which both the acoustic echo and the near-end speech are observed. On the other hand, echo cancelation applying blind source separation (BSS) can perform echo cancelation during double-talk, but its convergence speed is slower than that of the conventional adaptive algorithm. In this paper, a system that combines the conventional AEC and BSS is proposed. Simulation results show that the proposed system is effective for the echo-path estimation during double-talk and echo-path variation.