Library

Audio

Listen to a track, a tab, an input or the room, and drive any slider in the scene from what it hears

This layer draws nothing — it listens. It opens an audio input on the machine showing the scene and measures what is coming through it, four times a frame’s worth: an overall level, plus bass, mids and highs. Any slider on any other layer can then be tied to one of those, so a kick swells a lens, a hi-hat jitters a grid, a bassline breathes through a gradient. The input can be a microphone, or a virtual loopback device carrying whatever is already playing on the machine, which is how a track drives the picture. Nothing is recorded or uploaded; the sound is measured in the browser and thrown away frame by frame.