Improvement of accuracy of three-dimensional sound space synthesized by real-time "SENZI", a sound space information acquisition system using spherical array with numerous microphones

We proposed a sensing method of three-dimensional (3D) sound-space information based on symmetrically and densely arranged microphones mounted on a solid sphere. We call this method SENZI (Sakamoto et al., 2008). In SENZI, the sensed signals from each of the microphone is simply weighted and summed to synthesize a listener's HRTF, reflecting the listener's facing direction. Weighting coefficients are calculated for individual listeners based on their HRTFs. These coefficients are changed according to the listeners' head movement, which is known to provide important dynamic perceptual cue for sound localization. Therefore, accurate sound space information can be presented to unlimited number of listeners not only beyond the distance but also beyond the time. Recently we realized this method as a real-time system using a 252-ch spherical microphone array and FPGAs. By using this system, accurate sound space information up to around 10 kHz can be synthesized to any listeners. However, the SNR of microphones affected to the accuracy of synthesized sound-space information, especially under low frequency region. To avoid the effect, we used condition numbers as an index to synthesize accurate sound-space information in the low frequency region.