Sound made visible
A precision tone generator driving a real Chladni plate. Play a frequency and watch the grains find its nodal lines — the same physics as sand on a vibrating metal plate, computed live, sixty times a second.
Free · no account · installable · works offlineNot animation timed to a number — the actual signal, measured live and turned into the picture you're looking at.
Around 14,000 grains each step in proportion to the local vibration, so they collect on the nodal lines exactly as sand does. Measured: ~73% end up sitting on a node.
Each vertex tracks its own harmonic, read from the FFT. A sine moves one point; a square wave moves only the odd ones. The figure is Fourier's series, drawn.
Raise a frequency about 40 octaves and it lands in visible light. 440 Hz becomes 620 nm, an orange-red. Octaves share a colour, which falls out of the maths for free.
Load a file or open the mic and all of it follows that instead — pitch tracked to under half a hertz. Nothing is uploaded; it never leaves your device.
Bowls, wind chimes, church and temple bells, rain, thunder, ocean, mountains, snow and ice — all generated live. No samples, nothing to download.
Fullscreen, controls hidden, for an LED projector or a wall. Pair it with a real plate and a speaker if you want the physical version.
This is the part most frequency apps skip. If a number is measured, it says so. If it's popular but unproven, it says that too.
Schumann resonances, EEG bands, pulsar spin rates. Two millisecond pulsars are audible with no scaling at all.
Planetary tones from orbital periods, Kepler's 1619 interval ratios, the golden ratio, A432 and A440. Computed at runtime, so you can check them.
All 22 shrutis as exact just-intonation ratios, and gamelan slendro — flagged as an idealised model, because real gamelan is never standardised.
The modern solfeggio set, labelled honestly: a 20th-century invention, not medieval, with no verified physiological effect.
What OLI will not do. It is not a medical device and makes no medical claims. No audible frequency destroys cancer cells — the real technique runs near 1 MHz through a focused transducer, and a speaker is not a weak version of that. Those real numbers are included as reference entries that deliberately produce no sound. Sub-audible rates like the 7.83 Hz Schumann resonance are delivered as a binaural beat, never faked as a tone your speaker cannot make.