3D audio and ambisonics

Getting started working with spatial audio

What is it?

3D audio, or spatial audio, uses sound objects to place sounds around you in the environment. The effect is similar to gaming, where sounds appear to come from a particular direction.

Ambisonics captures sound in all directions, and allows it to be 'decoded' to play back on a variety of outputs. We render the sound as binaural in real-time, rotated according to the listener's orientation, ready to be played back on headphones. You can create ambisonics files using a special microphone to record the sound, or by spatialising sounds in a studio using software like Reaper, and then rendering to an ambisonic file as output.

ECHOES enables spatial audio by using its existing system to place sound triggers (or Echoes) which then play sounds (Elements) which have fixed GPS coordinates. These sounds are then spatialised around the listener, creating a 'real' environment in which you can approach and walk around sound objects.

How do I experience it?

To experience the full force of our spatial audio, you need headtracking. To understand more about headtracking, read here.

Support for 3D audio and ambisonics is available on both our iOS and android apps.

It will prompt you to use headtracking. Currently there are three options (two on android): 

  • Using the onboard compass (phone sensors). This will spatialise using the phone’s compass, so point the device where your head is pointing.
  • Using the Thingy:52 IoT bluetooth prototyping device. Attach it to a hat or the headband of a pair of headphones
  • Using the Apple Airpods Pro headphones, which support spatial audio (iOS only).

How do I create 3D sounds (sound objects)?

For 3D sounds, you can either:

  • Select 'Add a 3D sound element' when adding an Element. It'll have the '3D' switch turned on already, or
  • use your existing Elements: just enable the 3D button on the Element.

Each sound has its own 'marker' which indicates its virtual placement in real space. You can move these. Bear in mind that if the sound is physically far from you when you trigger the Echo, you likely won't be able to hear it.

Each sound will appear to come from its location. You can add multiple sounds per Echo, and have them at different positions, and these sounds will be spatialised around you, giving you the appearance of a 'sound' environment that changes as you move.

How do I upload ambisonic (360°) files?

For ambisonic sounds, select 'ambisonic' when creating an element.

Don't switch the type to ambisonic after creating a 'sound' otherwise it will fail when you try to play it in the app.

Once you've created an ambisonic Element, then you can upload a first, second, or third order, (ACN channel order, SN3D encoded) file (wav or aiff) and it will convert it.

We support HOA (2nd/3rd order). Just upload a 9 or 16-channel uncompressed file and the system will convert it. Bear in mind that these files are much larger, and using multiple, long files will end up with an enormous download.

Can I spatialise binaural sounds?

Binaural sounds are already encoded as spatial audio. You cannot rotate them or 'spatialise' them, much in the same way you cannot create a 3D model from a photograph.

If you use a binaural sound as a 3D 'sound object' the binaural effect will be lost and the sound will be played back as mono. For this reason it is best to use binaural sounds as they were intended: as stereo playback over headphones.