The videos that conform the experience are a mixture of 3D models and distorted audio files.
The 3D models are generated with photogrammetry. This means that elements of the physical world are scanned via a series of photos and then 3D modeled.
The distorted audio files are also recorded from the physical world and then altered by code.
The outcome of mixing these files is the animation of the 3D models reacting to the sound.
