Title :
Tracking multiple speakers using random sets
Author :
Vo, Ba-Ngu ; Singh, Sumeetpal ; Ma, Wing Kin
Author_Institution :
Dept. of Electr. & Electron. Eng., Univ. of Melbourne, Vic., Australia
Abstract :
Tracking multiple speakers in an acoustic environment involves jointly estimating the number of speakers and their states. This important problem in signal processing is challenging in theory as well as implementation. The paper presents a novel and fundamentally well-grounded framework for tracking multiple speakers using random finite sets. Simulations are also presented to demonstrate the performance in tracking a randomly varying number of speakers in a reverberant room.
Keywords :
acoustic signal processing; parameter estimation; reverberation; set theory; state estimation; target tracking; acoustic environment; finite sets; multi-target tracking; multiple speaker tracking; random sets; reverberant room; signal processing; state estimation; Acoustic applications; Acoustic signal processing; Acoustical engineering; Cameras; Filtering; Loudspeakers; Microphones; Monte Carlo methods; Particle tracking; State estimation;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on
Print_ISBN :
0-7803-8484-9
DOI :
10.1109/ICASSP.2004.1326268