Skip to content
Tinglissimo

ASMR and ambient sound, generated live

Eight sounds, one slider each, mixed live in your browser. Nothing is streamed and nothing is downloaded: every sound here is synthesised on your own machine as you listen, which is why it starts instantly, never repeats, and keeps playing with the network off.

Sound layers

Steady rainfall, with the droplets swelling and easing

A swell that rises and breaks about every ten seconds

Air moving past trees and edges, gusting slowly

A low roar with irregular crackles over it

Equal power at every frequency — the brightest hiss

Equal power per octave — softer than white, like a shower

Weighted heavily to the bottom — a deep, distant rumble

Two pure tones, one per ear. Headphones only — see the notes

Stopped. Choose your layers, then press play — sound only ever starts when you ask for it.

Every sound is generated on this page as you listen — no audio files, no streaming, and no requests to anywhere. Check this before you rely on it. This sound generator is provided free and without warranty, and its results are not professional advice.

Why it is called Tinglissimo

From pianissimo, the instruction written on a score when something is to be played very softly — and tingles, which is what the ASMR community has always called the sensation, because no ordinary word for it exists. Very softly, and tingles: that is the whole brief for this site in one word.

It is worth saying plainly what that does and does not promise. Nobody knows why ASMR works, whether it works for everybody, or whether the tingles and the sleep are the same effect. The honest position is that a lot of people find soft, continuous, unpredictable sound easier to fall asleep to than silence or than music, and that is what is being offered here — not a treatment for anything.

What you are actually listening to

There are no audio files on this site. Not compressed ones, not short ones — none. Each layer is built from noise generated a sample at a time in your browser and shaped by filters, the same way a synthesiser makes a sound rather than the way a player plays one back.

That is an unusual choice and it is worth explaining, because it is also the reason this works better than the alternative. A downloadable rain track is a few megabytes, which is a wait before anything happens and a repeat every thirty seconds once it does. The repeat is the problem: your ear finds the loop point within a couple of cycles, and once you have heard it you cannot stop hearing it. Generated sound has no loop point to find. The rain running now has never been produced before and will not be produced again.

The practical consequences: it starts in well under a second, it costs nothing in data after the page loads, it keeps playing on a plane or a train with no signal, and there is no copyright question attached to any of it. The maths that produces it is in sound.js, is about four hundred lines, and is tested against the physics it claims to model.

Building a mix that is worth listening to

Start with one bed and one detail. A bed is the broad continuous layer — rain, brown noise, ocean — and it should sit low enough that you stop noticing it after a minute. A detail is the layer that gives the bed somewhere to happen: crackle over rain, wind over ocean. Two layers is usually enough, three is plenty, and eight at once is mud rather than atmosphere.

Keep the total quieter than feels right at first. The point of a masking sound is that it stops being interesting, and anything you can still pick out after five minutes is a sound you are listening to rather than one you are covered by. If you cannot hear somebody knocking on the door over your mix, it is too loud — that is a useful test and it also happens to be the safe one.

The mixer never lets the layers add up to a clipped signal, whatever you do with the sliders. Adding a second sound scales the pair back by the square root of two rather than by half, because uncorrelated noise adds in power and not in amplitude, so turning up a new layer does not visibly duck the one you already had.

What sound can and cannot do

Continuous background sound helps some people fall asleep and helps some people concentrate. The mechanism that is best supported is dull and mechanical: it raises the noise floor, so the sudden changes that would otherwise pull your attention — a door, a car, a conversation through a wall — are smaller relative to what is already there. That is masking, and masking is a physical fact rather than a claim about your brain.

Everything past that gets weaker fast. ASMR — the tingling scalp response some people get from certain quiet sounds — is real, well documented, and simply absent in a large fraction of people, which is why one person finds a video transporting and the next finds it irritating. Binaural beats have a large popular literature and a thin research one. Nothing on this site is health advice and nothing here is offered as a remedy for anything.

What is fair to say is narrower and still worth something: if a steady sound makes your room less startling, it may make it easier to sleep in. That is the claim this site makes, and it is the only one it makes.

What each layer is made of

Generated by calling the same profile builders the mixer runs, so this table cannot drift from what you actually hear. "Movement" is the slow modulation that stops a layer sounding static.

LayerBuilt fromMovement
Rainbrown + pink + white noise, 3 voicesgain every 20s; gain every 9s
Ocean wavesbrown + pink noise, 3 voicesgain every 10s; gain every 10s
Windpink + white + brown noise, 3 voicesfrequency every 14s; gain every 9s
Fireplacebrown + pink noise, 2 voicesgain every 11s
White noisewhite noise, 1 voicesteady
Pink noisepink noise, 1 voicesteady
Brown noisebrown noise, 1 voicesteady
Binaural beattwo sine tonessteady

Every layer is synthesised from a pseudorandom sequence and a filter chain. There is no recorded audio anywhere on this site.

Common questions

Do I need to download anything to use this?

No. The page contains no audio files at all — the sounds are generated by your browser from a few hundred lines of maths, so the whole site is a fraction of the size of a single rain recording.

Does it keep playing when I am offline?

Yes. Once the page has loaded, the sound is produced locally and needs no connection. Turning off Wi-Fi mid-session changes nothing, which is not true of any streaming alternative.

Why does the volume start so low?

Because you may be wearing headphones and we have no way of knowing what your system volume is set to. The master starts at 20 per cent and every start fades in over about two seconds rather than beginning at full level.

Can I keep my mix for next time?

Yes. Slider positions and the timer setting are saved in your own browser and restored when you come back. Nothing is sent anywhere, and clearing your browser data removes them.

Other sounds and mixes