Robust speaker localization in a disturbance noise environment using a distributed microphone system

Ambient noises and room reverberations seriously degrade the accuracy of speaker localization in real environments. In this paper we describe the localization of a speaker supposing that the speaker utters under a disturbance noise with large amplitude. Our algorithm localizes a sound source by finding the position that maximizes the accumulated correlation coefficient between multiple channel pairs. After the estimation of the first sound source, a typical pattern of the accumulated correlation for a single sound source is subtracted from the observed distribution of the accumulated correlation. Subsequently, the second sound source is searched again. To evaluate the effectiveness of the proposed method, experiments of speaker localization were carried out in an actual office room. The result shows that average error distances of the multiple sound sources are less than 10cm.