Binaural Auralization of Microphone Array Room Impulse Responses Using Causal Wiener Filtering

Binaural room auralization involves Room Impulse Responses (BRIRs). Dynamic binaural synthesis (i.e., head-tracked presentation) requires BRIRs for multiple head poses. Artificial heads can be used to measure BRIRs, but BRIR modeling from microphone array impulse responses (RIRs) is becoming popular since personalized obtained any pose with low extra effort. We present a novel framework estimating signal signals, using causal Wiener filtering and polynomial matrix formalism. The formulation places no explicit constraints on the geometry of enables directional weighting estimation error. A noise model regularization balance filter performance gain. complete procedure RIRs also presented, employing proposed framework. An application example illustrates 19-channel spherical array. Direct reflected sound segments are modeled separately. compared measured shown waveform-accurate up at least 1.5\,kHz. At higher frequencies, correct statistical properties diffuse field components aimed for. listening test indicates small perceptual differences BRIRs. presented method facilitates fast data set acquisition use in dynamic viable alternative Ambisonics-based auralization.