Improve Spatial Audio Quality with the Universal HRTF

The spatial audio experience on Meta Quest is getting a big upgrade with the new Universal Head-Related Transfer Function, or Universal HRTF. This new function is the culmination of years of research and development, and it’s designed to help you easily deliver deeper immersion through improved localization and frequency accuracy. Dive in to learn more about this upgrade and how to improve spatial audio in your apps on Meta Quest.

An HRTF, or Head-Related Transfer Function, is a key component of spatial audio rendering. It helps us localize sound sources in 3D space and lets us differentiate between sounds that are above us, below us, in front of us, behind us, to the left of us, or to the right of us. Current Meta Quest apps and APIs use an HRTF model that was created from publicly-available data, and while it has served us well, we knew we could make it even better.

Thanks to Meta’s substantial investment in audio research, anechoic chambers, and measurement capabilities, we were able to capture our own data set consisting of nearly 150 high-quality HRTFs. We subsequently created a model that best represents the general population. This new model, the Universal HRTF, has proven to be far superior to the legacy model after extensive testing, and it translates to a better spatial audio experience for more people.

Thanks to the new Universal HRTF, people should have an improved spatial audio experience in two main areas: localization and frequency accuracy. Improved localization lets people more precisely detect where a sound is coming from within a given space, particularly when judging the elevation of sounds coming from above or below them. Improved frequency accuracy means that sounds will be much more natural, with less coloration and filtering.

Extensive user experience research with nearly 100 test participants has shown that the new Universal HRTF vastly outperforms the old model, both in subjective testing as well as objective performance measurements. The average subjective user preference score increased from 37 to 52 points. When asked to accurately locate a sound source in 3D space, test subjects performed better across the board, with elevation accuracy improving by 81% on average.

We’re launching the Universal HRTF in v55 of our Audio SDK as part of Presence Platform so developers can take advantage of this new capability and pass on the benefits to their communities. The Universal HRTF will replace the older HRTF model, so all you have to do is upgrade to v55 and recompile your app in order to reap the benefits. The Audio SDK currently supports Unity. In addition, Audiokinetic Wwise, Unreal Engine and FMOD will be supported in the coming months.