In the coming multimedia age, a seamless acoustic environment, in which two parties can talk to each other as if they were in the same room, regardless of the microphone positions, is desirable. We have developed a new high-performance acoustic echo canceller that creates such a seamless acoustic environment. It achieves fast convergence using the new exponentially weighted stepsize projection algorithm, high speech quality during double talk, and training-free operation through the use of voice-switching technologies.
|Number of pages
|NTT R and D
|Published - 1995
ASJC Scopus subject areas
- Electrical and Electronic Engineering