I am currently a research-only ARC QEII Fellow.
A list of my past courses can be found here: teaching history.
My research investigates signal processing related to spatial audio, models of the human auditory system, spatial audio perception, auditory scene analysis, and neuromorphic spike computation.
During my research, I invented a spatial hearing aid because conventional hearing aids do not enable spatial hearing, with significant disadvantage for social interaction. This invention allows modern digital hearing aids to provide acoustic cues for spatial hearing. Spatial cues are concentrated at the high-frequency end, the very part of the spectrum that is involved in most hearing loss; the algorithm transposes this information to lower frequencies without interfering with speech perception. A USyd spin-off company, VAST Audio , is commercialising the technology.
My current research focus is on spatial audio research. In 2005, I was awarded a QEII Fellowship to pursue this research which deals with all aspects of the spatial perception of sound. This includes the study of how humans localise sounds, the effect of room acoustics on sound perception, speech perception in rooms, recording of spatial sound fields, playback using our 32 spherical loudspeaker array, generation of augmented and virtual reality audio, 3D voice communication systems with location aware computing, and the role of spatial audio in music and cinematic presentations.
I have developed a dual, concentric spherical microphone array (SMA) for the recording, reproduction, and analyses of 3D sound. It uses a novel, physical construction which places microphones on a small, rigid sphere and also on a larger, concentric open sphere; this invention records 3D sound over a wider range of frequencies as dictated by the radii of the two spheres. We have developed new algorithms for beamforming and near-field acoustic holography that optimally combines modal coefficients across the two spheres obtaining better performance than existing arrays. Our current focus is on applying methods of convex optimisation to spatial sound field analysis and synthesis.