Brown vs White Noise

The widest gap on this site: −5.98 dB per octave against 0.00. Everything worth knowing about choosing between them follows from that one difference.

How Far Apart They Really Are

Both colours on this site were measured the same way: 120 seconds of output at 48 kHz, Welch's method, least-squares fit between 100 Hz and 10 kHz. Brown came out at −5.98 dB per octave. White came out at 0.00 dB per octave, flat to within ±0.08 dB. The interesting quantity is neither of those numbers on its own but the distance between them, which is 5.98 dB for every doubling of frequency.

That compounds fast, and compounding is what people underestimate. Anchor the two spectra together somewhere in the bass — 100 Hz is convenient — and the gap between them is about 20 dB by 1 kHz, about 32 dB by 4 kHz, and about 38 dB by 8 kHz. Thirty-eight decibels is not a shade of difference. It is the difference between a sound being present and a sound being absent.

There is a second measurement that makes the comparison meaningful rather than misleading. Both are normalised to −20 LUFS under ITU-R BS.1770-4 K-weighting, and all five colours here sit within 0.6 dB of each other. So switching between them does not change how loud the room is; it changes only where the loudness is spent. Brown spends nearly all of it under a few hundred hertz. White spreads it across the whole range, which — because frequency bands get exponentially wider as you climb — puts roughly half of white noise's total power in the single octave above 10 kHz.

Everything below is a consequence of those two facts. Same perceived level, radically different distribution.

The Rule That Decides Which One You Need

Auditory masking is frequency-specific. A sound raises your detection threshold for other sounds near it in frequency and does comparatively little for sounds far away from it. That is not a preference or a marketing position; it is how the cochlea resolves frequency, and it is the single most useful thing to know before choosing a noise colour.

Read practically, it means the question is never "which colour is better". It is "what frequency is the thing that keeps waking me up". Answer that and the choice makes itself.

Brown noise covers what is low. Traffic on a main road, an HVAC unit, a lift motor, a fridge compressor, footsteps on the floor above, the thump of a neighbour's music through a party wall. All of these live under a few hundred hertz, which is exactly where brown noise has concentrated its entire budget.

White noise covers what is high and what is broad. The consonants of speech leaking through a wall, a phone notification, a smoke alarm's low-battery chirp, birdsong at first light, a door closing, a dog. Speech is the important case: intelligibility is carried mostly by consonants in the 2–4 kHz region, and white noise has as much energy there as it has anywhere. Brown noise is roughly 30 dB down on white at 3 kHz, which is why raising the volume does not rescue it — you would need an absurd amount of bass in the room to lift the treble tail of brown noise into the consonant band, and long before you got there the low end would have become the new problem.

The corollary is worth stating because it saves people a lot of fiddling: if the two spectra have already told you the answer, no amount of A/B listening will overturn it. You cannot mask a conversation with brown noise, and you cannot mask a lorry with white noise, at any setting either of them will tolerate.

Your Speaker May Have Already Chosen for You

This is the practical consideration that gets skipped, and it overrules the spectra more often than not.

Phone speakers, laptop speakers and small Bluetooth pucks roll off steeply below roughly 200–300 Hz. That region is not a detail of brown noise; it is essentially all of it. Played through hardware like that, brown noise is not quiet — it is largely missing, and the instinctive fix of turning it up only amplifies the fraction the driver can produce. White noise, by contrast, has nothing important happening in the bass, so it arrives almost intact through the worst speaker you own.

So the honest first question is not which colour suits you but what you will be playing it through. On headphones or a speaker with a real driver, brown noise is a genuinely different experience and the comparison is fair. On a bedside phone, you are comparing white noise against a fragment of brown noise and calling the result a preference. If the phone is the plan and low-frequency intrusion is the problem, that is a hardware decision to make before it is a colour decision — and if the phone is the plan and you simply want something dark, green noise was built specifically to survive it.

One of Them Leaves the Room

Low frequencies pass through walls, floors and closed doors far more readily than high frequencies do. It is the reason you hear a neighbour's bass line but not their dialogue, and it applies to your noise generator exactly as it applies to theirs.

Brown noise at a level that solves your problem can become somebody else's. In a flat, a shared house, a room next to a nursery, or a partner who sleeps lighter than you do, this is a real constraint and not a fussy one. White noise is comparatively contained: high frequencies are absorbed by soft furnishings and stopped by an ordinary door, so it tends to stay where you put it.

Directionality cuts the same way inside the room. White noise is audibly directional — you can hear where the speaker is, and moving it further away changes the character as well as the level, because a room absorbs treble faster than bass. Brown noise is close to non-directional, so placement barely matters and the sound fills the space evenly. That is pleasant when you want to be enveloped and unhelpful when the person beside you would like to turn away from it.

Twenty Minutes Versus a Whole Night

Listening fatigue from broadband noise is overwhelmingly a high-frequency phenomenon, and this is where the two colours separate on duration rather than on function.

White noise carries its energy where the ear is most alert. That makes it effective, and it also makes it tiring in a way that builds over hours rather than minutes. A great many people who abandoned a white noise machine did so somewhere around week two, not on the first night. Brown noise has almost no treble to fatigue you with, which is the main reason it is the colour people run overnight, every night, for years.

The practical pattern that emerges from this is not a winner but a split: a bright colour for bounded sessions where you need something specific covered, and a dark one for the eight hours where you mostly need to not notice anything at all. Plenty of regular users of both end up doing exactly that without ever deciding it deliberately.

The Research Runs Entirely One Way — and Still Does Not Say What You Would Hope

There is an asymmetry here that almost nobody writing about brown noise mentions: every controlled study in this area used white noise or pink noise. Brown noise has never been the sound under test.

Nigg and colleagues (Journal of the American Academy of Child and Adolescent Psychiatry, 2024) pooled the trials of white and pink noise on laboratory attention tasks in young people with ADHD or elevated attention problems, and stated plainly that no studies of brown noise were identified. Not weak studies. None. Anything described as "brown noise, backed by research" is an extrapolation from a different sound, and given a 38 dB divergence at 8 kHz it is not an obviously safe one.

The temptation is then to conclude that white noise is the evidence-based choice, and that overstates it badly. On sleep, the relevant paper is Riedy, Smith, Rocha and Basner, Noise as a sleep aid: A systematic review (Sleep Medicine Reviews, 2021), which screened three databases and assessed 38 articles on continuous white noise and similar broadband noise. Following GRADE, it rated the quality of evidence for continuous noise improving sleep as very low — a conclusion the authors themselves note contradicts how widely it is used — and it raised the possibility that continuous noise may negatively affect sleep and hearing. That review is cited on noise websites as an endorsement roughly as often as it is read.

So the accurate summary of the evidence for this comparison is: one colour has been studied and the results are weak, the other has not been studied at all. That does not make either of them a bad idea. It makes the acoustics — which are settled — the only part of this page you should treat as established, and the clinical part something to hold loosely. Nothing here is medical advice, and persistent sleep trouble is a conversation for a clinician rather than for a noise generator.

On volume there is one measurement worth carrying away, and it is about hardware rather than colour. Hugh and colleagues (Pediatrics, 2014) measured 14 infant sleep machines at maximum volume: all of them exceeded 50 A-weighted decibels at 30 cm, and three exceeded 85 dBA, a level which sustained beyond eight hours would exceed adult occupational exposure limits. Whatever risk exists in this activity is a function of level and duration. It is not a function of which colour you picked.

A Test That Will Actually Tell You Something

Name the intrusion before you name a colour. Lie in the room at the hour that is the problem and work out what you are actually hearing. Rumble, machinery and thumps point one way; voices, chirps and birds point the other. Roughly half the people who ask this question discover they are trying to mask two different things and need to fix the louder one at source.

Compare them at the same setting, not at the same feel. Because both are normalised to −20 LUFS, the volume control is a fair comparison for once. Resist the urge to nudge one up because it seems quieter — that instinct is exactly the perceptual difference you are trying to measure.

Give each two consecutive nights. The first night with any new sound measures novelty. The second measures the sound.

Judge on wake-ups, not on liking. Which colour you prefer while awake and which one gets you through to morning are different questions with a habit of giving different answers. Brown noise wins the first far more often than it wins the second.

If both half-work, you probably want the middle. Pink noise sits between them at −3.07 dB per octave and is the sensible answer for anyone whose honest verdict is "white is too sharp and brown is too heavy".

The Verdict, Such As It Is

Neither colour wins, and any page that declares one the winner is telling you about its author rather than about acoustics. What the measurements support is narrower and more useful: brown noise for low-frequency intrusion and for long unbroken sessions, white noise for speech and sharp sounds and for shorter bounded ones — with the caveat that a small speaker quietly disqualifies brown noise regardless of what the spectra say.

Both are generated procedurally here rather than looped from a recording, so neither has a repeat point to start hearing at three in the morning. Free browser sessions run 30 minutes, which is enough to settle the comparison but not enough to run a night on. The free Android app removes the session limit and carries both colours plus a sleep timer, offline playback and background play — which is what you want once the comparison is settled and you simply need the sound to still be there at 4 am.

Frequently Asked Questions

Related pages

• Brown Noise for Sleep — The free web player, plus the full guide to brown noise for sleep, focus, and tinnitus.
• Brown Noise — The deepest colour, measured at −5.98 dB per octave — best for traffic rumble and long overnight sessions.
• White Noise — The flat reference colour, measured at 0.00 dB per octave — the best of the five at making speech unintelligible.
• Pink Noise — Equal energy in every octave, measured at −3.07 dB per octave — the natural-sounding default and the audio industry's calibration signal.
• Brown vs Pink Noise — 2.91 dB per octave apart — indistinguishable in the bass, nineteen decibels apart at the top, and one of them collapses on a phone speaker.
• White vs Pink Noise — Flat per hertz against flat per octave — two kinds of even, why the ear prefers the second, and why no trial has ever ranked the two.
• Brown Noise for Tinnitus — Masking versus habituation, why the pitch of your tinnitus decides the colour, and where sound therapy actually sits in the 2014 clinical guideline.