A spatial enhancement approach for binaural rendering of head-worn microphone arrays

Abstract

This paper builds upon a recently proposed spatial enhancement approach, which has demonstrated im- provements in the perceived spatial accuracy of binaurally rendered signals using head-worn microphone arrays. The foundation of the approach is a parametric sound-field model, which assumes the existence of a single source and an isotropic diffuse component for each time-frequency index. The enhancement approach involves the post-processing of an initial estimate of the binaural signals, in order to obtain a refined esti- mate of binaural signals which more closely represent the inter-aural cues corresponding to the sound-field model. In this contribution, the enhancement approach has been implemented as an open-source framework, written in both the MATLAB and C programming languages, and as a real-time audio plug-in. The frame- work was also extended to offer direction-dependent gain control of sound sources relative to the listener, and a frequency-dependent control of the direct-to-diffuse balance, which are modifications that may find application within future augmented reality headsets and assistive hearing devices.<br/>