Immersive Audio can be used to create seamless, fully three-dimensional aural environments that preserve the correct spatial sound localization for a group of listeners. Using a combination of human psychoacoustic principles and adaptive signal processing methods a new audio system has been designed that can provide a seamless immersive experience for the listener. This 10.2 channel system (by chris kyriakakis and tom holman) uses three front loudspeakers to reproduce the direct sound from the performance, and adds two wide loudspeakers to reproduce the reflections from the side walls. This results in increasing the sense of spatial envelopment. Three surround channels are used to seamlessly fill the space behind the listeners. Elevation information is reproduced from two height channels. The placement of the loudspeakers and the processing of the sound coming from them results in a combined sound that simulates natural hearing much better than today's surround sound systems.

One potential problem with multichannel systems is the lack of program material. While there are countless mono and two-channel stereo recordings, there are very few multichannel recordings in existence. A new method, called Virtual Microphones has been developed at the IMSC Immersive Audio Lab, that can synthesize microphone signals in various places in a hall from the information contained in just one reference signal. This method can be applied to existing stereo (or mono) recordings to synthesize the signals in several virtual microphones that can then be mixed with the existing channels to produce a true multichannel mix. Experiments in various venues, such as the Alfred Newman Recital Hall, will allow us to build a library of acoustical spaces that can be used to create multichannel immersive audio content.

home

synopsis

program a

program b

notes a

notes b

10.2

people

biographies

directions

poster

chew
kyriakakis
thurmond
april 2002