Abstract
We have considered new sound source location algorithms for one receiving point noticing a formant frequency. The automatic operation of the sound image localization is necessary for position estimation of a sound source in a large-scale building space like an auditorium. We measure sound source locations using the phase difference of a formant frequency between three microphones and the height of the receiving point. Normally the formant frequency has some fluctuation. Therefore, we propose two types of new sound source localization methods. One is to track the fluctuating formant component by selecting the maximum components of the fast Fourier transform (FFT) results of the bandpass filter outputs. Another is to estimate the fluctuating formant frequency by the zero-crossing detection and to apply to the Fourier transform (FT). Both methods use a echo avoidance method by the detection of the occurrence of the formant signal. From some experiments, we show that the proposed system can detect the stationary sound source location with the errors less than 0.6m and track the moving sound source under the echo environment.
