Logo: to the web site of Uppsala University

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

Accurate binaural rendering requires accurate reproduction of binaural signals at the eardrum, which in turn requires adequate binaural technology. We propose a method to measure head-related & headphone transfer functions with a two-microphone array in the ear canal. By implementing a cardioid directional pattern, the forward and reverse propagating sound pressure components are measured separately, thus avoiding the influence of standing waves in the ear canal on the measurements. The method is useful in filter design for individualized binaural rendering that, compared with the blocked-canal method, does not assume acoustically 'open' headphones to be used. The method also mitigates the excessive sensitivity to microphone position of regular open-canal measurements. Validation measurements are conducted using a natural scale replica ear and a MEMS microphone array.

Pre-processing of HRTF phase has proved useful to improve binaural rendering of order-limited spherical harmonics (SH) signals. The adjustment is typically applied at high frequencies and allows to reduce magnitude errors for directional sound field components. This article proposes a practical method for HRTF phase pre-processing using linear phase above a cutoff frequency, and a direction-dependent phase offset to maintain correct diffuse-field interaural coherence. Two applications are discussed - filter design for binaural rendering of microphone array or SH-signals, and reduced coloration in virtual source panning on virtual speaker setups. Factors influencing the perceptual transparency of the phase modification are evaluated subjectively and objectively.

This paper presents a comprehensive model for ear-signal level coloration in stereo amplitude panning, enabling the calculation of monaural correction filters that equalize the average coloration over a small area around the sweet spot. The model takes into account the speaker setup geometry, listener Head-Related Transfer-Functions (HRTFs), the employed pan-law, the direct-to-reflected sound ratio, and the correlation between the speaker signals at the listening position. Coloration in diffuse sound reproduction is also investigated. The coloration model is validated using binaural room impulse response measurements, and the correction filters are found to effectively reduce the difference in composite ear power spectrum between a discrete and virtual center source. A listening test on the perceived spectral difference between these two cases, with stereo setups in front of and behind the listener, indicate that the correction filter improves timbral similarity between a virtual and discrete center source for rear speaker panning. The test also indicates that remaining unmodeled coloration sources are large, especially for front panning. However, a second listening test finds that the correction filter improves accuracy of perceived direction in front panning by mitigating the phantom image elevation effect.