Title
Real-time speaker localization and speech separation by audio-visual integration
Abstract
Abstract| Robot audition in real - world should cope with motor and other noises caused b y the robot's own movements in addition to environmental noises and re - verberation This paper reports how auditory process - ing is improved by audio - visual integration with active movements The key idea resides in hierarchical inte - gration of auditory and visual streams to disam biguate auditory or visual processing The system runs in real - time by using distributed processing on 4 PCs con - nected by Gigabit Ethernet The system implemented in a upper - torso humanoid tracks multiple talkers and extracts speech from a mixture of sounds The per - formance of epipolar geometry based sound source lo - calization and sound source separation by active and adaptive direction - pass ltering is also reported
Year
DOI
Venue
2002
10.1109/ROBOT.2002.1013493
Robotics and Automation, 2002. Proceedings. ICRA '02. IEEE International Conference  
Keywords
Field
DocType
distributed processing,filtering theory,mobile robots,position control,real-time systems,robot vision,sensor fusion,speech recognition,adaptive filtering,audio-visual integration,direction-pass filtering,distributed processing,epipolar geometry,humanoid robot,multiple speaker tracking,real time system,reverberation,robot audition,sound source localization,sound source separation,speech separation
Noise,Computer vision,Visual processing,Reverberation,Epipolar geometry,Computer science,Speech recognition,Artificial intelligence,Hierarchical INTegration,Acoustic source localization,Mobile robot,Humanoid robot
Conference
Volume
Issue
Citations 
1
1
22
PageRank 
References 
Authors
2.09
8
4
Name
Order
Citations
PageRank
Kazuhiro Nakadai11342155.91
Ken-ichi Hidai29811.77
H. G. Okuno31308.45
Hiroaki Kitano43515539.37