Brown Noise for Sleep and Study: Does the Deep Rumble Help?
Does the rumble do anything?
Does a deep rumble make you sleep better, or does it only cover the sounds that wake you? The second half of that question has a mechanism behind it; the first is unproven, and so is most of the rest of what brown noise is sold for. When thirty-eight articles on continuous noise and sleep were graded together in a 2021 systematic review, the verdict on whether that sound improves sleep was very low quality evidence, with studies pointing both ways, including some in which constant sound disturbed sleep rather than helping it. So the useful claim is the modest one. A steady background sound raises the level a car door or a creaking floorboard has to stand above before it registers, and those abrupt changes are a common cause of waking.
The same mechanism, and the same limit on what can be claimed for it, is why brown noise turns up as often on a desk at four in the afternoon as on a bedside table at midnight. Both uses come down to covering an intermittent sound with a constant one. What differs is which intruders matter and how long the sound needs to run, and the second half of this article deals with the daytime case.
Where the rumble comes from
Brown noise is defined by its slope. Its power spectral density falls at six decibels per octave, against white noise’s flat spectrum and pink noise’s three. That number is the whole character of the sound, and its consequence is more lopsided than it looks.
Integrate that slope across the audible band and roughly nine tenths of the signal’s power sits below 200 Hz. By 200 Hz the level is already 20 dB down from where it started at 20 Hz; by 2 kHz it is 40 dB down. There is almost nothing left in the region where white noise hisses, which is why listeners who find white noise harsh at night often get on with brown noise instead, and why it gets compared to a distant waterfall or an aircraft cabin.
Brown, white or pink
| Noise color | Slope | Character at night |
|---|---|---|
| White noise | 0 dB/octave | Bright and hissy; strong masking across the band, but the treble can feel harsh |
| Pink noise | -3 dB/octave | Balanced; a common middle ground between the other two |
| Brown noise | -6 dB/octave | Deep and rumbling; gentlest on the ears, best matched to low-frequency intruders |
The slope also tells you what each one is good at covering. Masking works best where the masker has energy, so brown noise is the natural choice against traffic rumble, a humming fridge compressor or footsteps overhead. Against a high-pitched intruder it has little left to work with, and pink or white noise will do more.
The asymmetry is not quite total, and the exception has a name. Masking spreads further upward in frequency than downward: a low-frequency sound raises the threshold for tones above it considerably more than for tones below it, an effect Egan and Hake measured in 1950. The size of the spread depends on the masker’s level, and it grows with it — the upward skirt of a masking pattern broadens as the masker gets louder, and at low levels the pattern is roughly symmetrical. That matters here, because every other piece of advice on this page says to keep the level down. At a quiet overnight setting there is little upward spread to collect, so the effect is a footnote on the ranking above rather than a reason to doubt it: brown noise is not completely useless against a voice, but turning it up far enough to make the spread count would be the wrong way to cover one.
Using it overnight, and who it suits
- Keep it quiet. Masking needs to sit just above the intruding sound, not drown it. For context on the safe end of the scale, the National Institute on Deafness and Other Communication Disorders puts sounds at or below 70 A-weighted decibels as unlikely to cause hearing loss even after long exposure, with 85 dBA and above capable of causing it; normal conversation is 60 to 70 dBA. Overnight playback should be far below that, at the lowest level that does the job.
- Prefer a speaker to earbuds. Hours of in-ear playback irritates the ear canal and is easy to leave too loud. A small bedside speaker is the better overnight arrangement.
- Decide between a timer and all-night play deliberately. If what wakes you is a 3 a.m. delivery van, a 30-minute timer will not help. If the problem is only getting to sleep, all-night playback is unnecessary.
- Set it and leave the phone alone. Adjusting the sound in bed costs more than the adjustment gains.
The people most likely to benefit are those whose sleep is broken by intermittent low-frequency noise: city traffic, a neighbour’s plumbing, footsteps from the floor above, a snoring partner. It is also a reasonable next thing to try for anyone who has tested white noise and found it too bright. If the bedroom is already quiet and sleep is already fine, there is nothing for a masker to mask, and adding one is unlikely to change anything.
The same sound at a desk
For studying, the evidence is better than it is for sleep and it points the other way. Anyone deciding whether to switch brown noise on in a quiet room before an hour of reading should know that it will probably cost a little. That is the consequence of pooling 65 reading experiments in a Bayesian meta-analysis: comprehension was reliably, if only slightly, worse with sound in the background than without it, and plain noise was among the sounds that did the damage, not an exception to them. Speech the listener could understand and music with lyrics hurt more, which is the part that turns out to matter below.
What survives is the comparison that actually comes up. The choice is rarely noise against silence; it is noise against whatever the room is already doing. Replacing an intermittent, word-bearing distraction — a conversation in the next room, a television through a wall — with a constant, featureless one trades the worst category in that meta-analysis for the mildest. That is the case for a masker at a desk, and it is a narrower case than the usual claim that noise helps you concentrate.
Brown noise is a specific choice within it, and not the obvious one. Speech sits mostly above 200 Hz, which is where brown noise has already given up nine tenths of its power, so against a voice it is the weakest of the three colours and white or pink noise will cover more. Where brown noise earns its place by day is against low-frequency, repetitive machinery: a building site, a ventilation plant, a bass line from a flat below, a train line. It is also the colour that is easiest to tolerate for several hours at a stretch, which is worth something across a long afternoon even when a brighter colour would mask slightly better.
Giving a session a shape
A study session has a start and an end that a night of sleep does not, and the controls on this site are built around that. On /en/study-noise one press starts the sound, and a session timer counts down from a preset or a custom number of minutes; at zero an alarm plays and the sound stops with it. The Pomodoro button runs the same thing on a cycle, a 25-minute focus round followed by a 5-minute break, starting the next round on its own until it is switched off. With fade switched on, the sound ramps up at the start and down when the timer reaches zero, so the end of a block is audible without being abrupt.
Two practical notes for brown noise specifically. All five colours are generated at the same loudness, so switching from brown to pink mid-session changes the spectrum and not the volume, which makes an honest comparison possible without touching the level. Open the mixer at /en/mixer instead and brown can be layered underneath a little pink, which keeps the low weight while putting some energy back where speech lives — the combination that a room with both traffic outside and a conversation next door actually calls for.
Brown noise is a masking tool with an unusually low-heavy spectrum. That makes it comfortable to listen to for hours and well matched to the rumbling, intermittent noises that wake people up or interrupt an afternoon of work. Treat the stronger claims with suspicion — the review evidence for sleep benefit is weak in both directions, and for reading the measured effect of noise is a small cost rather than a gain — and judge it on the two questions you can answer yourself: do you wake up less often, and do you get further through the reading?
References
- Riedy, S. M., Smith, M. G., Rocha, S. & Basner, M. (2021). Noise as a sleep aid: A systematic review. Sleep Medicine Reviews 55, 101385 (evidence that continuous noise improves sleep graded very low quality; some studies report disturbance)
- Vasilev, M. R., Kirkby, J. A. & Angele, B. (2018). Auditory Distraction During Reading: A Bayesian Meta-Analysis of a Continuing Controversy. Perspectives on Psychological Science 13(5), 567-597 (65 studies; noise, speech and music all carry a small reliable cost; intelligible speech and lyrical music worst)
- Egan, J. P. & Hake, H. W. (1950). On the Masking Pattern of a Simple Auditory Stimulus. JASA 22(5), 622-630 (masking spreads further upward in frequency than downward)
- Colored noise: the spectral slopes of white, pink and brown noise — IEEE Technology Navigator
- Pink noise — ASA/ANSI Standard Acoustical Terminology
- White noise — ASA/ANSI Standard Acoustical Terminology
- Noise-Induced Hearing Loss — National Institute on Deafness and Other Communication Disorders (at or below 70 dBA unlikely to harm even after long exposure; 85 dBA and above can cause hearing loss; normal conversation 60-70 dBA)
Frequently Asked Questions
Is brown noise better than white noise for sleep
Neither is better in general. With a slope of minus six decibels per octave, brown noise keeps nearly all its energy at the bottom of the range and almost none in the treble, which is what makes it easier for many people to sleep through. White noise is flat and masks across the whole band. The right choice depends on what is waking you: brown noise suits low-frequency intruders like traffic and machinery, while a high-pitched noise needs a masker with energy up there.
Is it safe to play brown noise all night?
At a low level through a speaker, it is fine for most people. For scale, the NIDCD puts sounds at or below 70 dBA as unlikely to cause hearing loss even after long exposure, and normal conversation at 60 to 70 dBA; overnight masking should be far quieter than that. Avoid sleeping in earbuds for hours, which irritates the ear canal and is easy to leave too loud.
Does brown noise actually improve sleep quality?
Not on the evidence available. The 2021 systematic review that pooled this literature rated the case that continuous sound improves sleep as very low quality, and the underlying studies disagreed with each other, some finding constant sound disruptive. What remains defensible is narrower than an improvement in sleep quality: a steady background sound covers the abrupt noises that cause awakenings. That is an environmental fix, not a treatment.
Is brown noise good for studying?
It is a reasonable choice when the room is noisy and a poor one when it is quiet. Pooling 65 reading experiments, a Bayesian meta-analysis found comprehension reliably worse with sound in the background than without it — a small cost, but a cost, and plain noise was no exception to it. A masker earns its place only by displacing something worse, which is usually speech, the most disruptive category in that analysis. Brown noise is the weakest of the colours at that particular job, because speech sits above the band holding its energy; it belongs against machinery or traffic, while a conversation next door is better covered by pink or white noise.