Two-stage blind source separation based on ICA and binary masking for real-time robot audition system

We newly propose a real-time two-stage blind source separation (BSS) for binaural mixed signals observed at the ears of humanoid robot, in which a single-input multiple-output (SIMO)-model-based independent component analysis (ICA) and binary mask processing are combined. SIMO-model-based ICA can se...

Full description

Saved in:
Bibliographic Details
Published in2005 IEEE/RSJ International Conference on Intelligent Robots and Systems pp. 2303 - 2308
Main Authors Saruwatari, H., Mori, Y., Takatani, T., Ukai, S., Shikano, K., Hiekata, T., Morita, T.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.01.2005
Subjects
Acoustic signal processing
binary masking
Blind source separation
Ear
Humanoid robots
ICA
Independent component analysis
Interference
Microphones
Real time systems
Robot audition
Signal processing
Source separation
Online AccessGet full text

Cover

More Information
Summary:We newly propose a real-time two-stage blind source separation (BSS) for binaural mixed signals observed at the ears of humanoid robot, in which a single-input multiple-output (SIMO)-model-based independent component analysis (ICA) and binary mask processing are combined. SIMO-model-based ICA can separate the mixed signals, not into monaural source signals but into SIMO-model-based signals from independent sources as they are at the microphones. Thus, the separated signals of SIMO-model-based ICA can maintain the spatial qualities of each sound source, and this yields that binary mask processing can be applied to efficiently remove the residual interference components after SIMO-model-based ICA. The experimental results obtained with a human-like head reveal that the separation performance can be considerably improved by using the proposed method in comparison to the conventional ICA-based and binary-mask-based BSS methods.
ISBN:0780389123
9780780389120
ISSN:2153-0858
2153-0866
DOI:10.1109/IROS.2005.1544983