| Paper: | AE-P1.1 | ||
| Session: | Loudspeaker and Microphone Array Processing | ||
| Time: | Wednesday, May 17, 10:00 - 12:00 | ||
| Presentation: | Poster | ||
| Topic: | Audio and Electroacoustics: Loudspeaker and Microphone Array Signal Processing | ||
| Title: | Blind Speech Separation Using PARAFAC Analysis and Integer Least Squares | ||
| Authors: | Kleanthis Mokios, Nicholas Sidiropoulos, Alexandros Potamianos, Technical University of Crete, Greece | ||
| Abstract: | We propose a new two-step frequency domain algorithm for blind speech separation (BSS) for unknown channel order. This new approach employs parallel factor analysis (PARAFAC) to separate the speech signals and a novel integer-least-squares-based method for matching the arbitrary permutations in the frequency domain. The proposed algorithm offers guaranteed convergence and good separation performance, measured both quantitatively and in subjective tests. Performance gains in signal-to-interference ratio of up to 10 db are achieved for certain source-sensor geometries. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012