SMPTE Next Gen Audio
June 2016 – Video was not the only topic of discussion at the SMPTE ETCA event. Audio was also a key part of the conversations and discussions. Sound creates a great deal of the emotion of a visual experience. It can both, enhance and soften a scene as well as direct the point of attention in a scene.
The new visual standards of higher resolution and color, as well as 360degree experiences, are advancing their associated sound to improve the immersion along with the improved image.
There is a bandwidth challenge for multi-channel sound. Just as more pixels takes more bandwidth and bigger data sizes for the video, as audio moves from stereo to higher orders of multi-channel sound, the data scales also. This is true for broadcast or reception at home, business or mobile (automotive) platforms and locations. A current theater standard that is making its way to automobiles and small theatres is the 22.1 surround sound. This is a 2 plane sound environment that uses and object based hybrid channel system to produce 3D sound with spatial positioning based on metadata. The object based methodology and the accompanying metadata help reduce the amount of data required to deliver this sound.
Because of the mobile community and new headworn device immersive 3D sound is being developed that can be delivered through binaural headphones. The systems being developed are bringing theater level sound accuracy to the headphone system by having the contents remixed to 5.1, 7.1 or Dolby Atmos sound. These solutions are currently widely used in both console and PC gaming and VR applications.
One of the challenges in remixing these multi-channel systems, if you have to know what you want it sound like in the end, in order master it to come out that way. The tools and the codecs do not simply deliver the positional sound with the correct result, they have to be mixed that way by someone who has the intent and the timing for where to place the sounds.
Qualcomm has been working on integrating into their hardware solutions a codec that addresses higher order ambisonics or HOA. This is also known as MPEG-H and is sometimes referred to as Scene-based audio. The scene-based audio format has been designed to represent the audio landscape or scene as a field of pressure values at all points in a given space, over time. This sound pressure based representation has been engineered to be an absolute representation of the 3D sound-scape. The encoding is designed to be independent of the equipment used to record and playback the sound.
