Free online grey noise generator
This grey noise generator plays perceptually flat grey noise directly in your browser — no download, no account, and no audio streamed from a server. Press play and the signal is synthesised live on your device; adjust the level with the slider until it sits where you want it.
Grey noise (spelled gray noise in American English — the same thing) is the colour designed around human hearing rather than around a measurement meter. Where white noise delivers equal power to every frequency and consequently sounds bright, grey noise is shaped by an equal-loudness contour so that it sounds evenly weighted from bottom to top.
What grey noise is
The ear is not equally sensitive across the spectrum. Sensitivity peaks in the 2–5 kHz region, where speech consonants live, and falls away steeply at the low end and gently at the very top. Equal-loudness contours — the curves standardised in ISO 226, descended from the classic Fletcher–Munson work — describe that shape.
Grey noise takes a random broadband signal and applies the inverse of that curve. Frequencies you hear poorly are boosted, frequencies you hear acutely are attenuated, and the result is noise that reads as uniform to a listener even though a spectrum analyser shows a distinctly curved response. It is the only common noise colour defined by perception rather than by a spectral slope.
Grey noise vs white noise
| Grey noise | White noise | |
|---|---|---|
| Defined by | Equal loudness to the ear | Equal power per hertz |
| Measured spectrum | Curved — dipped in the midrange, lifted at the extremes | Flat |
| Perceived character | Even, full, comparatively soft | Bright, hissy, midrange-forward |
| Typical use | Hearing tests, tinnitus masking, psychoacoustics | General masking, broadband signal testing |
A useful way to hear the difference is to switch between the two at a fixed volume in the full noise generator. White noise appears to concentrate in the upper-mid region; grey noise spreads out and feels less like it is aimed at one part of your hearing. Compare it with the white noise generator side by side.
Grey noise vs pink noise
Pink noise is the other candidate for "natural sounding" noise, and the two are often confused. Pink noise falls at a fixed 3 dB per octave, giving equal energy per octave, which happens to approximate the long-term average spectrum of music. Grey noise follows a curve rather than a straight slope, and that curve is derived from hearing thresholds, not from music.
In practice pink noise still has audible low-end weight, while grey noise trades some of that for a more even distribution of perceived loudness. If you are calibrating a room, use pink noise. If you are trying to mask a specific internal sound or run a perceptual test, grey is the better baseline.
What grey noise is used for
- Tinnitus masking and retraining therapy: a masker that does not itself draw attention to one frequency region is easier to tolerate for long sessions.
- Audiology and hearing tests: perceptual uniformity is the point when you are measuring a listener rather than a loudspeaker.
- Psychoacoustic experiments: grey noise gives a neutral perceptual backdrop against which test stimuli can be judged.
- Everyday masking for sensitive listeners: people who find white noise too sharp and brown noise too muffled frequently land on grey.
- Long focus sessions: the reduced midrange emphasis makes it less tiring than white noise over several hours.
Download grey noise as a WAV file
To keep a copy for offline use, open the noise generator with grey selected, set the duration, choose a sample rate between 22.05 and 96 kHz and a bit depth of 16, 24 or 32-bit float, then export. The file is generated locally and downloaded as an uncompressed WAV, so it imports into any DAW or editor and loops seamlessly.
How this generator makes grey noise
A true grey noise generator applies a full inverse equal-loudness filter. Harmonoise uses a lightweight approximation of that shape: a pink noise core, produced with a multi-stage filter bank, blended with a proportion of white noise and scaled. The result reproduces the perceptual character of grey noise — less midrange emphasis than white, more high-end presence than pink — at a fraction of the processing cost, which matters when everything runs in a browser tab.
For casual listening, masking, and focus work the difference from a full ISO 226 filter is not audible. For formal audiometric measurement, use calibrated laboratory equipment rather than any web-based generator, including this one.
Frequently asked questions
Is it grey noise or gray noise?
Both spellings refer to exactly the same signal — "grey" is the British spelling and "gray" the American one. There is no technical distinction, and audio literature uses the two interchangeably.
What is the difference between grey noise and white noise?
White noise is flat when measured: equal power in every hertz of bandwidth. Grey noise is flat when heard: an equal-loudness curve is applied so the frequencies your ear is least sensitive to are boosted and the ones it is most sensitive to are pulled back. White noise sounds bright and hissy; grey noise sounds comparatively even across its range.
What is grey noise used for?
Mainly hearing science and tinnitus work. Audiologists and psychoacoustic researchers want a masker that is perceptually uniform rather than mathematically uniform, so that no part of the spectrum dominates the listener's attention. Outside the clinic, people who find white noise harsh and brown noise muffled often settle on grey.
Can I download grey noise as a WAV file?
Yes. The full noise generator exports grey noise as an uncompressed WAV at up to 96 kHz and 32-bit float, in mono or stereo, with the length set to the millisecond. The file is rendered in your browser and saved directly to your device.