Dirac VR – More technical information


 
 

Dynamic HRTFs

A major barrier to the truly immersive AR/AR experience is eliminated with the introduction of Dirac VR™, which features patent pending dynamic HRTFs (head-related transfer functions) that account for the isolated movements of the head in relation to the human torso.

If a phone rings to your left, the sound waves enter your left ear a fraction of a second before your right ear, then bounce off your head and shoulders in various ways, allowing your brain to localize the sound and instinctively rotate your head to face the source. Normally, you just turn your head while your torso remains still. This is the reality.

Yet, conventional HRTF models still assume that the head and torso move in tandem. These are static HRTFs. Although there exists a number of public HRTF databases, as of yet, not one of them correctly models the dynamic movements of the head in relation to the torso. Such failure to accurately reproduce dynamic head movements in virtual reality is one of the reasons why existing positional sound solutions don’t quite sound realistic.

Dirac VR’s Dynamic HRTF measurement captures the dynamic HRTFs resulting from isolated head movements, while the sound object remains in a fixed position, with high precision and high resolution. Both localization and sound quality improve significantly when we take into account the exact position of the head relative to the torso.

 

High-Resolution, Efficient HRTF Measurements

The torso-head dilemma isn’t the only immersion limitation Dirac VR resolves. Existing databases of static HRTFs lack resolution and suffer from measurement noise and errors. Such shortcomings result in inaccurate positioning of audio sources, and sound quality degradations. Dirac VR utilizes a highly sophisticated HRTF measurement process that, when compared to existing approaches, is exponentially freer from noise and error.

In order to create a more immersive audio experience, current state-of-the-art technologies rely on the individualized HRTF measurements of each listener — a time-consuming and laborious process which ultimately lacks scalability. This reflects the current industry belief, that individual HRTFs are necessary for attaining high-quality positional audio with headphones.  However, we prove that with the right approach, this is really not the case. A minimal improvement can be obtained, but it is so minuscule that it's hardly noticeable. 

When Dirac VR’s high resolution measurement process is combined with dynamic HRTFs, the benefits of individualization become virtually immaterial and ultimately obsolete. Most importantly, the combination ensures the reproduction of much more accurate, natural, and positional sound without coloration—resulting in a truly immersive experience for the user.  

 

Natural and Realistic Three-Dimensional Space

In virtual reality, if someone calls out from behind you in a church, with the right audio cues not only will you hear the sound coming from that direction, you’ll also recognize that you are truly standing in a church (not a bathroom or a long corridor). The reason for this is that each space has its own specific acoustical behaviors that color the sound, helping you identify where you are.

In order to simulate a sound coming from a particular direction in a particular space, we need to precisely model the direct wave, the early reflections, diffuse reverberation, and the impact made on the sound by the body, head, and ears. To achieve this—apart from the accurate rendering of the head-related transfer function (HRTF)—Dirac VR uses a proprietary 3D reverberation engine. The 3D reverberation engine realistically renders the critical aspects of a room’s unique acoustics to create a sense of a specific space and distance which mirrors reality without undermining the sound quality. Content developers can even utilize this tool to design acoustic characters of specific spaces and sound with a high degree of flexibility in their virtual soundscape designs.

Using the same platform, Dirac engineers have successfully recreated the acoustical behaviors of the Gothenburg Concert Hall from inside a Volvo XC90—the winner of “Best Audio System over £25,000” at the inaugural Car Tech Awards in 2016. 

 

Patented Headphone Optimization Technology

When used in combination with Dirac HD Sound, our patented digital headphone optimization technology Dirac VR creates the ultimate immersive virtual reality experience.

Regardless of how well designed they are, all headphones suffer from one fundamental problem: They are mechanical devices. And all mechanical devices produce acoustical imperfections. A fact which is detrimental to virtual reality, where any acoustical shortcoming by the headphones can distract the user and, ultimately, destroy the entire illusion.
Using leading frequency-domain and time-domain correction algorithms, Dirac VR enhances the audio performance of headphones by minimizing any coloration attributable to the headphones themselves. As a result, a more balanced and transparent sound with richer details is achieved.  

 

Efficient Implementation for Low Memory and CPU Consumption

As the leading provider of audio post-processing algorithms for the smartphone industry, we have a unique know-how when it comes to low power implementations of high-end audio technology. The processing and models used for Dirac VR have been optimized for extremely low CPU and memory requirements, and high-resolution minimum-latency tracking of micro head-movements. The technology is optimized for battery-limited mobile devices, while at the same time enabling the highest possible sound performance.