White Noise for Babies

White noise is a reasonable thing to try and a poor thing to run loud: the NHS lists gentle background noise among the things that may help distract a crying baby, and the only measurement anybody has published on these machines is a caution rather than a clearance. Fourteen infant sleep machines were played at maximum volume and every one of them was above the noise limit set for infants in hospital nurseries at thirty centimetres. Volume, distance and duration are the whole subject. Quiet, across the room, and for the settling rather than the whole night.

Those two halves belong on one page, because the genre keeps them apart. The pages selling a machine leave out the measurement; the pages quoting the measurement rarely admit that plenty of babies settle better with sound than without it. Both are true at once, and the answer sits in the middle.

What is actually supported

The NHS lists having some gentle noise in the background among the things that may help distract a crying baby, alongside holding her close, moving gently, rocking and a warm bath. That is the whole of the published support, and it is worth reading it for what it says: noise may distract a baby who is already crying. It is not a claim about how long she sleeps.

Nobody this site is allowed to cite has measured how much sleep continuous sound buys a baby, at what age, or for how long. What families report is consistent and unmeasured: a steady sound settles some babies and masks the door, the stairs and the older child, so fewer things wake her. That is a plausible mechanism rather than a finding.

There is also a trap in judging it yourself. The NHS notes that crying tends to increase until a baby is around two weeks old and then gradually reduces from about three months, so anything introduced at six weeks is introduced on the way down that slope. If you want to know whether it is doing anything for your baby specifically, run a fortnight with it and a fortnight without and write the nights down — a printed night sheet filled in with one hand beats a memory formed at four in the morning.

The one measurement anybody has published

One team measured the sound of fourteen infant sleep machines played at maximum volume, at three distances: 30, 100 and 200 centimetres. It is the only study of its kind, and worth going through in detail, because it is the entire evidence base for the safety half of this question.

All fourteen exceeded 50 A-weighted decibels at 30 centimetres — the recommended noise limit for infants in hospital nurseries. Thirty centimetres is not a strange distance to test at. It is a machine clipped to the cot rail, or sitting on the mattress by her head, which is where these things routinely end up.

Three of the fourteen produced levels above 85 A-weighted decibels. At that level, played for more than eight hours, the exposure exceeds the limits set for accumulated noise exposure in adults at work, and it carries a risk of noise-induced hearing loss. Eight hours is not a stress test either — it is a night.

Two things the study did not do, said plainly. It did not follow any baby and find hearing damage; it measured what the machines can produce. And it did not test them at the setting you actually use. Every figure above is at maximum volume, which is why the honest reading is about capability rather than about your bedroom: these machines can be run loud enough to matter, and the controls that decide whether yours is are on your side of the room.

Why nobody can hand you a number

Search for how loud to set it and you will be given a figure. Nobody publishes one as a target. There is no safe decibel level for a nursery published by the NHS, the American Academy of Pediatrics, the NICHD or any peer-reviewed paper this site may stand on, and the 50 dBA above is a hospital standard the machines exceeded rather than a level anybody has cleared for a night at home. A page handing you a number to set your machine to has invented it.

The same goes for any app, this one included. A volume slider on a phone is a proportion of whatever the device is already set to, played through a speaker of unknown size into a room of unknown shape. That is why the white noise player on this site labels its range in words rather than decibels, opens quiet and defaults to a timer. It cannot hear your room, so it does not pretend to.

The three levers you actually have

Volume, distance and duration are the three things the study varied, and they are the three things you control.

Turn it down until it is background rather than foreground. If you have to raise your voice over it, it is too loud: you should be able to talk in the doorway in a normal voice.

Move it away. Distance is the cheapest fix in the whole subject, because sound falls off sharply with it — that is precisely why the study measured at a metre and two metres as well as at thirty centimetres. Across the room, on a chest of drawers or a shelf, pointing at a wall rather than at her head. Never on the cot rail, never in the cot, never under the mattress. A speaker, a phone and a cable are all things that do not belong in there: babies sleep on their back, on a firm flat mattress, in a cot clear of blankets, pillows, bumpers, soft toys and anything with a wire on it, in your room for at least the first six months.

And shorten it if you can. The hearing risk in the study came from a level sustained across more than eight hours, so a sound that covers the settling and then fades sidesteps most of the question without giving up the part that helps.

All night, or a timer

Plenty of families run it all night, and the argument for doing so is real: if the sound is masking a road, a flat above you or a sibling who gets up at six, then switching it off at the point she falls asleep removes the thing that was working. If you run it all night, run it quiet and far away, and take the maximum setting off the table entirely.

The argument for a timer is that it removes the duration lever from the equation and costs you very little. Thirty minutes with a slow fade covers the settling; if she surfaces at two in the morning and needs it back, you have learned something useful about what it is doing. Either choice is defensible. What is not defensible is loud, close and all night at once, which is the combination the measurement was warning about.

Which colour, and why it matters least

White noise carries equal energy at every frequency and sounds hissy. Pink weights the low end and sounds like steady rain. Brown weights it further and sounds like distant surf. Nobody has compared them in babies, so any page ranking them is ranking its own preference.

The reason families drift towards the deeper ones is worth stating, because it is about volume rather than taste: the harsh high frequencies in white noise are the part that makes it feel loud, so a pink or brown sound at the same setting is easier to leave quiet. Choose whichever you can bear at a low volume, and spend the attention you saved on where the speaker is instead.

How to stop using it

There is no hurry unless it has become a problem for you — a nursery that does not use it, a holiday, a room you cannot control. If you do want to stop, take the volume down a little every few nights over two or three weeks until it is barely there, then switch it off; or shorten the timer by five minutes every few nights until it is doing nothing. Both work, and both are duller than they sound.

Pick an ordinary fortnight for it. A week with teeth, a cold or a house move in it is a week when everything you change gets blamed for everything that happens, and you will end up unable to tell which was which. If you are already travelling, the travel and jet lag guide makes the opposite case — a familiar sound is one of the few things you can carry into a strange room.

When the crying is not a noise problem

Sound is a settling aid, not an explanation. If she cannot be settled at all, is unusually sleepy or hard to rouse, is feeding poorly or not gaining weight, or her cry does not sound like her usual one, that is a call to your pediatrician or health visitor rather than a volume question — and so is a worry that will not go away. If the nights themselves are the problem rather than the settling, the day and night confusion guide and the bedtime routine guide cover more of the ground than a machine can, and safe sleep basics covers what goes in the cot. This page is general information, not medical advice, and it cannot see your baby.