Evaluation of the extension and coloration of multiple listening zones synthesized by the shared field reproduction system

抄録

A conventional sound field synthesis re-creates a highly accurate 3D sound space. However, it is difficult to present spatial sound to multiple listeners because an impractical number of loudspeakers is required for a single synthesized field to cover them all. The present research introduces a new approach, termed ``shared sound field synthesis,'' to multi-zone reproduction, which allows multiple users to simultaneously listen to the spatial sound corresponding to a single sound field. The proposed method considers a new kind of multi-zone transfer function in the spherical harmonic domain to characterize the output of each loudspeaker in the reproduction array. These are calculated using a phase compensation formula to account for sound propagation between zones. Numerical simulations show that the proposed method can consistently achieve a two-zone reproduction with a consistent distortion level below -6 dB and is stable with respect to the relative position between the listening zones, except when the zones and target source are aligned. The size of the low distortion is similar to that of the single-zone synthesis obtained through high-order Ambisonics. Moreover, the proposed method dose not proved significant coloration at low frequency or near the center of the listening zones.