Interaural cross-correlation: Sources of variability in concert halls

Abstract

Impulse responses were measured inside multiple concert halls. Data analysis identified the level of the direct sound relative to the level of the architectural reflections as a significant source of IACC80 variability. Excluding the direct sound from the IACC80 integral durations (IACC80×DIR) significantly decreased the variability. The remaining variability seemed to relate to the arrival direction of the architectural reflections. A testing model capable of various configurations and a scale measurement method were developed. Models that had reflections approaching the receiver from the sides produced much lower IACC80×DIR values than those without reflections approaching the receiver from the sides. The various configurations of architectural elements (i.e., balconies and stage canopy) and surface treatments (i.e., diffusive and specular) affected IACC80×DIR values to a lesser extent than altering the arrival direction of the architectural reflections. It was concluded that the general direction from which the architectural reflections arrive and the finish of the architectural surfaces are both sources of IACC80×DIR variability (the former having a much greater effect). It was also concluded that unless the direct sound is excluded from the integral duration, variability of interaural cross-correlation values does not relate as strongly to the arrival direction of the architectural reflections. [Work supported by the Concert Hall Research Group and the University of Florida.]