Head-related transfer function personalization for the needs of spatial audio in mixed and virtual reality

Virtual Reality (VR) and Mixed reality (MR) devices are typically a head mounted screens, aiming to project objects in an entire environment (VR), or in addition to the existing environment (MR). In both cases, these devices should have the means to create the spatial audio images of the projected objects. Considering the compact form factor of the devices, the most common approach is using binaural audio reproduction through a pair of headphones, isolated (VR) or acoustically transparent (MR). One of the key factors for successful realization of such spatial audio system is personalization of the user Head-Related Transfer Functions (HRTFs). In this paper, we will present and compare several approaches for doing so under the heavy constraints of the VR/MR devices design.