White Noise and Ambient Sound Generator

How to use the white noise generator

1. Set the three Noise layers. By default only Pink noise is on, at 60 %. Move White noise or Brown noise up to add them to the mix, or down to 0 % to turn them off.
2. Adjust the Master volume. It starts at 40 %: start low, especially with headphones, and raise it slowly.
3. Choose a Sleep timer: 15 minutes to 8 hours, or Off (play until stopped). Pick how long the Fade out at the end lasts: 10 s, 30 s, 1 min or 5 min.
4. Press Play. The sound fades in over a second and a half and the remaining time is shown under the button. Press Stop at any time.

You can move every slider while the sound plays; changes are smooth. Changing the timer or the fade while playing restarts the countdown from the new value. When the timer runs out, the sound fades to silence and stops by itself.

The noise is synthesised by your browser with the Web Audio API. Nothing is downloaded or streamed, so it works offline once the page is loaded and uses very little data.

White, pink and brown noise: what’s the difference?

All three are random signals; they differ in how their energy is spread across frequencies.

• White noise has the same energy in every hertz. Because an octave at the top of the spectrum (10–20 kHz) contains far more hertz than one at the bottom, it sounds bright and hissy, like an untuned radio or a TV between channels.
• Pink noise loses 3 dB per octave, so every octave carries the same energy. Our ears perceive it as balanced and softer, close to steady rain or wind in leaves. This tool makes it with Paul Kellet’s well-known filter, accurate to a fraction of a decibel across the audible range.
• Brown noise (also called red or Brownian noise) loses 6 dB per octave. The low end dominates: it sounds like a distant waterfall, surf, or the hum inside a plane cabin. Here it is made by a “leaky” integration of white noise, which keeps it from drifting.

Each layer is normalised to the same average level, so the sliders feel comparable. Layering works well: a little white noise over brown noise adds some “air” to a deep rumble.

Sleep, focus and masking

A steady noise masks sudden sounds (traffic, a neighbour’s TV, a snoring partner) because a change in level is what wakes us or pulls our attention, not the sound itself. Many people also use noise to concentrate in open offices or to calm a baby. Research on sleep and noise is mixed: it helps some people fall asleep faster and does nothing for others. If you use it for a baby, keep the device far from the cot and the volume low.

The sleep timer exists so the sound doesn’t run all night. A long fade-out avoids the jolt of an abrupt stop, which can itself wake you up.

Tips and limits

Each colour is a 12-second loop. The end of every loop is cross-faded into its start, so there’s no click, and the layers start at different points so their loops never line up. Nobody can hear a repeat in random noise of this length.

The browser keeps the sound playing when you switch tabs, but a phone may pause it when the screen locks, depending on the system. For the best result, plug the phone in and leave the page open.