A new comprehensive 3D sound-space sensing and reproduction method by a symmetrically and densely arranged microphone array

摘要

Sensing of precise and comprehensive sound-space information is an important issue to realize highly-realistic communications. However, conventional methods are not capable of sensing precise and comprehensive sound-space information at a listening point so that the information can be reproduced for many remote listeners individually at the same time. We invented a new method to sense 3D sound-space information based on symmetrically and densely arranged microphones. The proposed method are classified into two types: one is based on a spherical microphone array consisting of hundreds of microphones called SENZI (Symmetrical object with ENchased Zillion microphones) and the other is a modified dummy-head called ASURA (A Symmetrical and Universal Recording Array). Both types are symmetrical so that they are responsive to listener's head motion. Sensed signal from each of the microphones is simply weighted and summed to synthesize a listener's HRTF. This method is expected to comprehensively sense 3D sound-space information in accordance with the listener's head motion, who is listening at a remote place. An important advantage of the proposed method is that comprehensive 3D soundspace information is acquired without knowing the sound source positions. We examined the performance of the proposed method in terms of the accuracies of synthesized HRTFs as a function of the number of microphones and for various microphone array geometries by computer simulations. In this lecture, we introduce the basic concept and the performance of the proposed method. The results of performance analysis indicate that our proposed method can sense 3D sound-space with high precision.