Development of 3D sound generation system for multimedia application

A system aiming at implementing 3D sound, so called, 3DSGS (3D Sound Generation System), is introduced. This system supplies spatial attributes of a sound source which occurred in a particular place to other dull sounds faithfully, which is made possible by utilizing HRTF (Head Related Transfer Function) as convolution filter to modify spectral characteristics of the input source. The 3DSGS developed consists of two modules through which one can realize 3D sound images for 8 positions and 20 directions around a listener with ease. The technique of generating 3D sound image with only S/W implementation and the newly measured HRTF set using Neumann Dummy Head are described in detail.