One layer per microphone. Vertical threads mark a chord: two places holding the same frequency in the same moment.
Open microphones, permanently installed in places around the world, are streaming right now. This piece listens to several at once — chosen to be as far apart as possible — and composes from whatever happens to coincide. Everything you hear is arriving live. Nothing was recorded in advance.
Shapes, not samples.
One layer per microphone. Vertical threads mark a chord: two places holding the same frequency in the same moment.
Distances are to the nearest microphone already playing — how much a place would actually add. Every one of these is someone's installation.
Faders move a place as a whole — the live stream and the layers the piece generates from it. None of this touches the analysis: pull one to nothing and the spectrogram, the pattern matching and the narration carry on. Level matching sits upstream of the faders: over a few seconds it lifts the quiet places a long way and trims the loud ones a little, toward a common level. Each strip shows what it is doing, in dB. Turn it off to hear the set as it really arrives.
Everything below advances on its own — the microphone list is fetched, the furthest-apart set is chosen, and each stream is opened as it is reached. There is nothing to configure and nothing to click through.
Waiting for the first gesture — browsers will not start audio without one.
These are live installations maintained by the people credited in the sidebar and relayed by the Locus Sonus soundmap. Some are in quiet places at quiet hours — silence is the material, not a fault. Start at a low volume.