White Noise for Babies: How Loud, How Long, and Is It Safe?

The womb is loud. Researchers have measured sound levels inside it at roughly 70 to 90 decibels, somewhere between a vacuum cleaner and a lawnmower. Your baby spent nine months surrounded by the whoosh of blood flow, the thump of your heartbeat, and the muffled noise of your day. For a newborn, a silent nursery is the unfamiliar thing.

The second job white noise does is masking. Babies pass through a light sleep phase roughly every 45 minutes, and in light sleep a sudden sound can wake them: a door closing, a sibling shouting, a dog two gardens over, a car outside. A steady background sound smooths those spikes so they never register. This is why white noise earns its keep in flats, terraced houses, and any home with a toddler who does not understand the concept of a sleeping baby.

Now the honest bit about the evidence. The most-quoted study, by Spencer and colleagues in 1990, found that 80% of newborns fell asleep within five minutes with white noise playing, against 25% in silence. That is a striking result, but it was a small study and it is now decades old. Research since then has been supportive without being dramatic: continuous background sound seems to help many babies settle faster and wake less, and the effect varies from baby to baby. White noise is a cheap, low-risk aid that helps a lot of families. It is not a guarantee, and some babies take no notice of it at all.

The guideline you will see everywhere is about 50 decibels at your baby's ear. That figure comes from recommendations on how loud hospital nurseries should be, and the American Academy of Pediatrics applies the same number to sound machines at home. The World Health Organisation uses a similar figure for extended infant exposure. No UK health body has issued its own limit, so 50dB at cot distance is the sensible number to work to.

50dB is quieter than most parents expect. It is roughly a quiet conversation at home, or light rainfall on a window. There are two easy ways to check you are near it:

• The talk test. Stand next to the cot with the sound playing and speak at your normal volume. If you can be heard comfortably, the level is about right. If you would need to raise your voice, turn it down.
• A free decibel meter app. Phone apps are rough tools rather than lab equipment, but they are accurate enough for this job. Hold the phone at the cot, roughly where your baby's head lies, and watch the average over a minute or so. You want it reading around 50 or below.

Why this matters: a 2014 study in Pediatrics tested 14 popular infant sound machines at maximum volume and found every one of them exceeded 50dB at every distance tested. Some passed 85dB when measured 30cm away, close to the level at which long exposure damages adult hearing. The lesson is simple and calm: never run a machine at or near maximum, and keep it well away from the cot. Used at a sensible volume and distance, white noise is one of the lowest-risk sleep aids there is. If you are choosing a machine, our white noise machine roundup flags which models have a volume range that goes low enough.

Across the room. A shelf, chest of drawers, or windowsill at least 2 metres from the cot is the standard advice, and it solves the volume problem almost by itself, because sound falls away quickly with distance.

Never put a sound machine on the cot rail, inside the cot, or on the mattress. At that range the volume at your baby's ear can be far higher than the machine sounds from where you are standing, and anything in or attached to the cot is a problem in its own right under safer sleep guidance: a clear cot means clear of everything.

Two practical points. If your machine is mains-powered, route the cable behind furniture and well out of reach, because a loose cable near a cot is a strangulation risk. And if you use your phone instead of a machine, the same rules apply: face down, plugged in, on aeroplane mode, 2 metres away.

Both are reasonable choices, and families land on each for good reasons. Here are the trade-offs.

Running it all night covers the light-sleep phase your baby passes through every 45 minutes or so, when a sudden silence can wake them just as surely as a sudden sound. It also masks the 5am bin lorry and the neighbour's alarm. The cost is more hours of exposure, which is why the volume rule matters most for all-night use: at 50dB or below there is no evidence of harm from running it for the whole sleep, but the lower you can set it while it still does its job, the better.

A timer means fewer hours of sound, which some parents simply prefer. The catch is the cut-off. If the machine clicks off while your baby is in light sleep, the change can cause the very wake-up you bought it to prevent. If you use a timer, set it long enough to cover falling asleep plus at least the first sleep cycle, and if your baby keeps waking around the time it switches off, that is your answer: go continuous.

There is no hidden right choice here. Pick the one that suits your baby and your household, and change it if the evidence in front of you says to.

It is a sleep association, and it is one of the easiest ones to live with. Compare it with associations that need your body, like feeding to sleep or contact napping: white noise works the same whether you are in the room or in bed asleep yourself, it never gets tired, and it travels in a coat pocket via a portable machine or a phone app.

Plenty of adults sleep with a fan or rain sounds every night and never think of it as a problem to fix. The same logic applies to your baby. If white noise helps everyone sleep, you can keep using it for as long as you like. There is no age at which it must stop and no medical reason to set a deadline.

The one scenario worth watching is if you find yourself turning it up over time to get the same effect. Louder is not the fix, and creeping volume usually means something else is driving the waking. The last section of this guide covers the usual suspects.

You might want to stop because nursery or the childminder cannot replicate it, because packing for travel is one job too many, or because you would rather your child could sleep anywhere. All fair. Going cold turkey usually backfires, so do it gradually:

1. Turn the volume down a small amount every two to three nights.
2. Over two to four weeks, bring it down to barely audible.
3. Switch to a timer, around 45 minutes, so it covers falling asleep and then stops.
4. Shorten the timer over a week or two, then drop it altogether.

Most families get through this in two to four weeks without much drama. If sleep falls apart partway, turn it back up, hold steady for a few weeks, and try again. Nothing is lost by waiting, because there is no deadline.

The colours describe how the sound's energy is spread across frequencies, which in practice just means how bright or deep it feels to your ear.

• White noise has all frequencies at equal strength. It sounds like a bright static hiss, or an untuned radio. It is the most studied type for infant sleep and the best at masking sharp, sudden sounds.
• Pink noise turns the treble down. It sounds like steady rainfall or wind in trees, and many parents find it softer on their own ears. There is some adult research linking it to deeper sleep, and the infant evidence is thin but positive.
• Brown noise goes deeper still, a low rumble like heavy rain on a roof or a distant washing machine. It has had a huge moment on social media with adults. There is very little infant-specific research on it.

Which should you use? Whichever one your baby settles to. There is no strong evidence that any colour beats the others for baby sleep, and the same volume and distance rules apply to all three. What matters far more is that the sound is continuous and monotonous. Skip anything with built-in variation, like crashing waves, birdsong with pauses, or lullabies that swell and fade, because the changes in volume act as little prods during light sleep.

White noise fixes one thing: a room where sound is waking your baby. When the waking is coming from somewhere else, no volume setting will touch it. The usual suspects are worth ruling out before you blame the machine:

• Temperature. A room that is too warm or too cold wakes babies reliably. The Lullaby Trust recommends 16 to 20°C; our room temperature guide covers how to check and what to adjust.
• Light and timing. A 5am wake in June is usually about dawn light and the shape of the day. Our early waking guide walks through it.
• Hunger, illness, teething, or a developmental leap. All of these override any sleep environment, and all of them pass or get fixed by other means.
• Overtiredness or a schedule that has drifted. A baby who is wired at bedtime needs the day adjusted rather than the room.

One more reassurance, stated once and calmly. At the volumes and distances described in this guide, there is no evidence that white noise harms babies' hearing. Every baby in the UK is offered a hearing check in their first weeks through the NHS newborn hearing screening programme. If you ever have a separate worry about your baby's hearing, such as not startling at loud sounds or not turning towards your voice as the months go on, raise it with your health visitor or GP. That is what they are there for, and hearing concerns are always worth a proper look rather than a guess.