Headphone-delivered three dimensional sound in NPSNET

Abstract

The current MIDI-based sound system for the distributed virtual environment of NPSNET can only generate aural cues via loudspeaker delivery in two dimensions. To further increase the sense of immersion experienced in NPSNET, a sound system is needed which can generate aural cues via headphone delivery in three dimensions. The approach taken was to explore the different feasible methods of rendering and presenting headphone-delivered spatial sound. One alternative was to implement a sound server capable of the real-time rendering of three dimensional sounds. Another alternative was to create a library of pre-recorded positioned sound files. In software, new algorithms were developed to integrate the sound server into NPSNET and to provide a table lookup capability for NPSNET's new spatial sound file library. The result of this research is a sound server capable of rendering up to twenty-four simultaneous sounds for a single participant in NPSNET using off-the-shelf sound equipment and computer software. This sound server was tested during numerous demonstrations of NPSNET. This research provided another method of increasing a participant's level of immersion in NPSNET through the use of aural cues.