2006 IEEE International Conference on Multimedia and Expo
Sound Source Separation of Trio using Stereo Musig Sound Signal Based on Independent Component Analysis
Toronto, ON, Canada
July 09-July 12
ISBN: 1-4244-0366-7
Satoru Morita, Faculty of Engineering, Yamaguchi University, 2-16-1 Tokiwadai, Ube, Japan
Yasuhito Nanri, Faculty of Engineering, Yamaguchi University, 2-16-1 Tokiwadai, Ube, Japan
It is necessary that the number of the observation signals equals to the number of source signals, if independent component analysis is used to perform the sound source separation. It is difficult to perform sound source separation from a stereo music sound signal when the number of sound sources are more than two. We propose the technique to perform sound source separation from a stereo music sound signal that the number of sound sources is more than two using the frequency analysis and independent component analysis.
Citation:
Satoru Morita, Yasuhito Nanri, "Sound Source Separation of Trio using Stereo Musig Sound Signal Based on Independent Component Analysis," icme, pp.185-188, 2006 IEEE International Conference on Multimedia and Expo, 2006