Title :
A new algorithm for joint blind signal separation and acoustic echo canceling
Author :
Schobben, Daniel W E ; Sommen, Piet C W
Author_Institution :
Eindhoven Univ. of Technol., Netherlands
Abstract :
The problem of joint blind signal separation and acoustic echo cancelling arises in applications such as teleconferencing and voice controlled machinery. Microphones pick up a signal of the desired speaker together with contributions of other speakers and loudspeakers in these applications. The contributions of these loudspeaker signals to the microphone signals need to be cancelled. The remaining signals are then separated so that the individual local speakers are recovered. In this paper an extension of the previously introduced convolutive blind signal separation algorithm, CoBliSS is presented. This extended algorithm is capable of performing combined blind signal separation and acoustical echo cancelling at a low computational cost. The performance of the extended CoBliSS algorithm is evaluated using audio that is recorded in a real acoustical environment
Keywords :
acoustic signal processing; audio signal processing; computational complexity; convolution; echo suppression; loudspeakers; microphones; teleconferencing; acoustic echo cancellation; audio signals; convolutive blind signal separation algorithm; extended CoBliSS algorithm; hands free telephony; joint blind signal separation; loudspeaker signals; loudspeakers; low computational cost; microphone signals; microphones; performance; real acoustical environment; teleconferencing; voice controlled machinery; Acoustic applications; Australia; Blind source separation; Loudspeakers; Machinery; Microphones; Signal processing algorithms; Symmetric matrices; Teleconferencing; Uniform resource locators;
Conference_Titel :
Signal Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth International Symposium on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
1-86435-451-8
DOI :
10.1109/ISSPA.1999.815814