Suround sound and soundfield manipulation
development of signal processing techniques for the real-time manipulation and control of multichannel audio
”Researchers
- Bruce Wiggins
- Peter Lennox
The professional audio industry has recognised the potential of emerging media formats, such as DVD, to carry encoded multichannel audio data. This has led to a number of research opportunities in the development of signal processing techniques for the real-time manipulation and control of multichannel audio.
The teaching team within the Division of Electronic Systems already has experience working with multichannel sound formats and in particular spatial localisation processing for multichannel speaker arrays. Members of the group are actively researching into models, methods and techniques for the adaptive correction and alignment of multichannel reproduction equipment and the development of a DSP based distributed processor adaptive decoder architecture for the real-time alignment, manipulation and reproduction of multichannel sound formats.
Many attempts have been made at multi-channel sound reproduction such that a listening environment can be extended from simple stereo to horizontal surround (2D) and full periphonic (3D) representations. Examples of this, in the past, have included Quadraphonics, Ambisonics and Dolby Surround systems. Recently greater attention has been given to the psycho-acoustic principles behind multi-channel sound systems. Current work in the area of psycho-acoustics suggests that a number of parameters, in addition to those originally suggested by Gerzon, may need to be considered when designing multi-channel sound encoders and decoders. However, at present no consensus of opinion exists which can encompass the different psycho-acoustic effects into a single workable system. This has led to great diversity in the subjective quality of multi-channel audio reproduction systems. Furthermore, given the confusion over psycho-acoustic effects, and their assessment, it is difficult to determine an optimal method for capture and subsequent reproduction of full periphonic sound fields.
In this research project, a comprehensive survey of psycho-acoustic principles has been carried out. Particular emphasis has been placed upon the influence psycho-acoustic effects have on the perception of sound sources within a multi-channel listening environment. This has led to the design of a powerful, yet flexible multi-channel sound platform and listening
environment that has enabled a HRTF based perceptual model to be used in the quantitative investigation into three dimensional sound field recording and reproduction.
It is this model that will facilitate the design and implementation of a coherent, hierarchical, multi-channel system capable of psycho-acoustically accurate sound field reproduction over headphones or any number of transducers arranged in any configuration. An accurate, three-dimensional recording method will also be designed, based around the Soundfield MKV capsule arrangement, which will address problems with the current state of the art multi-channel recording systems.
