Brown Noise
A deep, low-frequency rumble — closer to a waterfall heard through a wall than to static. Measured on this player at −5.98 dB per octave.
What Brown Noise Actually Sounds Like
Brown noise is the darkest of the common noise colours, and most people reach for water to describe it: a strong waterfall, a river in flood, surf heard from the top of a cliff. Others hear machinery instead — the cabin drone of a long-haul flight at cruise, a tumble dryer two rooms away, wind pushing steadily against a window at night. What every one of those has in common is that the detail sits low. There is body and weight to the sound, and almost no hiss on top of it.
Put side by side with white noise the difference is not subtle. White noise sounds like it is arriving from a point somewhere in front of you — bright, papery, a little abrasive. Brown noise sounds like it is coming from the room itself. That enveloping quality is why it is the colour people most often pick for a full night rather than a twenty-minute session, and why listeners who bounced off white noise years ago tend to get on with it immediately.
The player above generates brown noise continuously rather than looping a recording, so there is no repeat point at all. Nothing to notice at three in the morning, nothing to start listening for once you have heard it a few hundred times.
The Spectrum, Measured Rather Than Quoted
Brown noise — also called Brownian noise or red noise — has a power spectral density that falls by 6 decibels for every doubling of frequency. Halve the frequency and the power quadruples. That is the textbook definition, and nearly every page written about brown noise repeats it. Very few of them have checked whether the sound they are serving actually does it.
This one was measured. Over 120 seconds of output at 48 kHz, analysed with Welch's method and fitted by least squares between 100 Hz and 10 kHz, the slope came out at −5.98 dB per octave, deviating from a straight line by no more than ±0.36 dB across that whole range. That is the −6 dB per octave ideal, within a third of a decibel.
In practical terms it means the energy is stacked hard at the bottom. By 1 kHz — roughly the region where the vowels of ordinary speech live — brown noise has already shed about 20 dB relative to 100 Hz. Above about 5 kHz there is very little left at all. That single fact explains almost everything brown noise is good at, and everything it is bad at.
One more measurement worth stating, because it affects how the site feels rather than how it looks on paper: all five colours here are normalised to −20 LUFS using ITU-R BS.1770-4 K-weighting, with only 0.6 dB of spread between them. Equal-energy noise is not equal-loudness noise, so without that step brown would sound almost inaudible next to blue. The left and right channels are also generated from independent random sequences (measured correlation around 0.005), so the stereo image is genuinely wide rather than one mono signal copied to both ears.
What Brown Noise Is Genuinely Good At
Low-frequency intrusions. Traffic rumble, an HVAC unit, a lift motor, the bass from a neighbour's music, footsteps on the floor above. Auditory masking works best when the masker and the unwanted sound occupy the same part of the spectrum, so a low-frequency problem needs a low-frequency masker. Brown noise is the only colour on this site with real power below 100 Hz, and for these specific complaints nothing else here comes close.
Long sessions. The absence of high-frequency energy is the reason people can leave brown noise running for eight hours without the sound starting to grate. Listening fatigue from broadband noise is largely a treble phenomenon; brown noise simply has no treble to fatigue you with.
Listeners who dislike white noise. If you have tried a white noise machine and described it as sharp, tinny, or like a hiss inside your head, brown noise is the obvious next thing to try. It is the same idea executed at the opposite end of the spectrum.
Layering. Because it sits so low, brown noise stacks underneath rain, music or a fan without fighting them for the same frequency range. In the Android app this is what sound mixing is for — brown noise as a floor, with rain or thunder on top of it.
What Brown Noise Is Bad At
It cannot mask anything high. A smoke-alarm low-battery chirp, a phone notification, birdsong at first light, the sibilance of speech leaking through a wall — all of these sit in a region where brown noise has essentially nothing left to offer. If the thing keeping you awake is a conversation rather than a lorry, white or green noise will do more for you than brown will, no matter how loud you set it.
It needs hardware that can reproduce it. Phone speakers, laptop speakers and small Bluetooth pucks roll off steeply below roughly 200–300 Hz, which is exactly where most of brown noise's energy lives. Played through them, brown noise is not merely quiet — a large part of it is missing. The instinctive response is to turn the volume up, which only raises the fraction the speaker can actually produce. Headphones or a speaker with a real driver change the experience so completely that it is worth testing before deciding brown noise is not for you.
It travels. Low frequencies pass through walls, floors and doors far more readily than high frequencies do — the reason you hear a neighbour's bass line and not their dialogue. If you live in a flat or share a wall with a light sleeper, brown noise is the colour most likely to become somebody else's problem.
Sustained low-frequency sound does not suit everyone. Some listeners describe a sensation of pressure or fullness in the ears after a while. If that is your experience, pink noise gives you a similar unhurried character with a good deal less weight at the bottom, and green noise removes the sub-bass region entirely.
What the Research Does and Does Not Show
The honest position on brown noise is that there is almost no research on it specifically, and pages that imply otherwise are overselling.
The most relevant systematic review is Riedy and colleagues, Noise as a sleep aid: A systematic review (Sleep Medicine Reviews, 2021), which examined continuous white noise and similar broadband noise as a sleep aid across 38 articles. Its conclusion was not the ringing endorsement it is often quoted as: the evidence that continuous noise improves sleep was judged to be of very low quality, and some of the included studies reported no benefit or worse sleep. Nothing in that review isolated brown noise as a condition.
The other frequently cited work is about different colours and different outcomes altogether. Nigg and colleagues (Journal of the American Academy of Child and Adolescent Psychiatry, 2024) pooled the trials on white or pink noise and laboratory attention tasks, and found a small benefit for people with ADHD or high ADHD symptoms — and no benefit for anyone else. Once again, brown noise was not the sound under test.
What is uncontroversial is the acoustics. Auditory masking is settled physics: a sound raises the threshold at which you can detect other sounds close to it in frequency, and a −6 dB per octave signal is unusually well suited to masking low-frequency household noise. Whether reliably masking noise translates into measurably better sleep is a different question, and it has not been answered well for any colour — and has barely been asked for this one.
So: if brown noise works for you, use it. It costs nothing, it is not a drug, and at a sensible volume it has no obvious downside. Just treat any page that describes it as clinically proven — including any app store listing — with scepticism. Nothing here is medical advice, and a sleep problem that persists is a conversation to have with a clinician rather than with a noise generator.
Getting the Most Out of It
Set the volume by the intrusion, not by the noise. Turn it up only until the sound you are trying to cover stops being distinct, then stop. Past that point extra volume buys you no extra masking — the intrusion is already gone — and simply adds exposure. A useful rule of thumb: if you could still hold a normal conversation over it without raising your voice, you are in a sensible range.
Placement matters less than usual. Low frequencies are far less directional than high ones, so a speaker on the floor, behind a chair or on the far side of the room works about as well as one beside your head. This is the opposite of the advice you would give someone using white noise, where the source position is very audible.
All night or on a timer. The web player generates sound continuously and free sessions run for 30 minutes, which is enough to fall asleep to but not enough to run overnight. The Android app removes the session limit and adds a sleep timer, offline playback and background play, which is the setup most people end up wanting once they have decided the sound suits them.
How It Compares With the Other Colours
Brown noise sits at one extreme of the spectrum of noise colours, and the useful way to think about the others is as a slider from dark to bright. Pink noise at −3.07 dB per octave is the halfway house: still warm, but with the detail brown noise throws away. White noise is flat, bright and by far the most studied. Green noise concentrates on the middle and drops both extremes, which makes it the one colour that sounds much the same on a phone speaker as on headphones. Blue noise tilts the other way entirely at +3.08 dB per octave, and is best understood as an engineering signal rather than a sleep sound.
If your problem is rumble, start here. If your problem is voices, start with green or white. If you are not sure, pink is the safest default.
Frequently Asked Questions
Why does brown noise sound so much deeper than white noise?
Because of how the power is distributed. White noise carries equal power in every hertz, which puts an enormous share of its energy in the top octaves. Brown noise sheds 6 decibels of power for every doubling of frequency, so by the time you reach the treble there is almost nothing left. The measured slope of this player is −5.98 dB per octave, which is why it reads as a rumble rather than a hiss.
Is brown noise the same thing as red noise or Brownian noise?
Yes — three names for one signal. "Brownian" refers to Brownian motion, the random walk named after the botanist Robert Brown, because integrating white noise produces exactly that kind of drift and exactly this spectrum. "Red" is an analogy with light, where red sits at the low-frequency end of the visible spectrum. The colour brown has nothing to do with it.
Will brown noise mask my neighbour's television?
Partly, and probably not in the way you want. Brown noise will cover the low-frequency body of the sound — the thump of music and effects, the general presence of a voice. What it will not cover are the consonants, which live well above where brown noise has any energy left, and consonants are what make speech intelligible. If half-audible dialogue is the specific thing keeping you awake, green or white noise will do more.
Why does brown noise sound so quiet on my phone speaker?
Because most of the signal is below the range a phone speaker can physically produce. Small drivers roll off steeply below roughly 200–300 Hz, and brown noise puts the bulk of its energy underneath that. You are not hearing quiet brown noise so much as a fragment of it. Headphones, or any speaker with a real bass driver, will sound dramatically different. If headphones are not an option, green noise is deliberately built to avoid this problem.
Can I leave brown noise playing all night?
The sound itself is generated continuously with no loop point, so there is nothing that degrades or becomes noticeable over a long session. The practical limits are the web player's 30-minute free session and your device staying awake; the Android app removes the session limit and adds a sleep timer and offline playback. As with any sound played for hours, keep the level modest — duration and level determine exposure, not the colour.
Is there scientific proof that brown noise helps you sleep?
No, and it would be misleading to say otherwise. The 2021 systematic review in Sleep Medicine Reviews that covers continuous broadband noise as a sleep aid rated the evidence as very low quality, and no trial in it tested brown noise specifically. What is well established is the underlying acoustics of masking. Plenty of people find brown noise helps them; that is a reasonable basis for trying it, and not the same thing as clinical proof.