Acoustic Lab Fraunhofer

New audio format from creators of MP3 promises next-generation surround sound

Image credit: Fraunhofer

The Fraunhofer Institute has been demonstrating its next-generation audio format, MPEG-H, at MWC and is hoping its advanced capabilities will see it become the new de facto standard for broadcasters and creators of virtual-reality (VR) experiences.

The Institute, which was part of the team that first created the now ubiquitous MP3 file format, is seeing increasing adoption of MPEG-H around the world.

It allows content creators to create true surround-sound audio experiences. While 5.1 surround sound is commonplace, MPEG-H effectively has no limit to the number of audio sources it can replicate and can even emulate sound in a 3D space by aping both vertical and horizontal sources.

In addition, it can apply audio to individual objects in virtual environments, a significant boon to creators of VR content.

While demoing the technology, Fraunhofer product manager Julien Robilliard explained that many VR experiences only provide simple stereo audio and fail to track the user's head and adapt the virtual soundscape accordingly.

Virtual objects can each be assigned a location and the sound emitting from them changes depending on whether they are behind, in front, or to the side of the user. The experience even works while using a normal pair of stereo headphones.

For broadcasters MPEG-H brings a number of benefits, allowing them to manipulate the volume of many different audio layers.

“Take film dialogue for example,” Robilliard said, “you can raise the volume of dialogue in a movie without affecting the effects.

“If you’re watching sports you can turn off the commentary and just experience what it’s like to be in the crowd in the stadium.”

In a Formula 1 demo, Fraunhofer showed that the level of all of these different elements could be adapted according to user preference and they could even silence the sound of everything except what was being said by the driver’s in-helmet microphone.

Like MP3, MPEG-H has also been designed to be supported on a wide range of devices.

“A lot of cost can be saved because you’re mixing once and the same content will adapt to the device whether it’s being played back on a mobile phone, on a TV, or a VR device, because it will always adapt,” he said.

“The TV will know that it only has stereo speakers so the decoder will down-mix all of these objects for stereo output.

“Now if you have an AVR with 7.1 setup the decoder will unpack to make best use of those speakers. It will always adapt to the end device.”

The format has also been adapted for efficiency in much the same way that MP3 allowed songs to take up less virtual space while still providing similar audio quality (fight me audiophiles).

Truly immersive audio can be achieved at roughly the same file size as 5.1 did in the past, a necessity for TV that has limited data broadcasting capability or mobile devices with small internal storage drives.

Korean TV networks have already started broadcasting in MPEG-H for major sporting events, although support is currently limited to a handful of TVs made by Samsung and LG.

“The standard was finalised over three years ago but the biggest part of the work that needs to done now is to get people to implement it,” Robilliard said.

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles