Computational Auditory Scene Analysis and Its Application to Robot Audition

Authors:
Hiroshi G. Okuno;Tetsuya Ogata;Kazunori Komatani;Kazuhiro Nakadai
Affiliations:
Kyoto University;Kyoto University;Kyoto University;Honda Research Institute Japan
Venue:
ICKS '04 Proceedings of the International Conference on Informatics Research for Development of Knowledge Society Infrastructure
Year:
2004

Citing 0
Cited 2

Easy Living in the Virtual World: A Noble Approach to Integrate Real World Activities to Virtual Worlds

WI-IAT '09 Proceedings of the 2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Volume 02
Bathroom activity monitoring based on sound

PERVASIVE'05 Proceedings of the Third international conference on Pervasive Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We are engaged in research on computational auditoryscene analysis to attain sophisticated robot (computer) humaninteraction by recognizing auditory awareness. Theobjective of our research is the understanding of an arbitrarysound mixture including non-speech sounds and musicas well as voiced speech, obtained by robotýs ears (ormicrophones embedded in the robot). The main issues aresound source localization, separation, and recognition atsignal processing levels, and signal-to-symbol transformationat the interface level to symbol processing levels. Thelatter is critical in developmental communication and weare developing an automatic onomatopoeia recognition system.This paper overviews our activities in robot audition,in particular, active direction-pass filter (ADPF) that separatessounds originating from a specific direction by integratingsound source localization and visual processing.ADPF is implemented on three kinds of robots anddemonstrates separating and recognizing three simultaneousspeeches with a pair of microphones.