If you want a single number to work from, it is 30 decibels — the World Health Organization's long-standing guideline for continuous background noise inside a bedroom over an eight-hour night.1 A fridge two rooms away will beat it. So will a partner turning over, a car door on the street, or a boiler firing at 4am. This guide explains what a decibel actually measures, how loud everyday sounds really are, what nighttime noise does to your sleep, and which of the available fixes are worth your money.
What counts as noise?
Noise is sound you did not ask for. That definition sounds glib, but it is the one that matters for sleep, because the physics of a sound tells you nothing about whether it will wake you. Two people can lie in the same room, exposed to identical sound pressure, and only one of them stirs.
Sound itself is straightforward enough. A vibrating object — a speaker cone, a vocal cord, a lorry's engine — pushes and pulls on the air around it, creating travelling bands of slightly compressed and slightly rarefied air. Those pressure fluctuations arrive at your eardrum, which moves with them, and your brain interprets the movement as sound. No medium, no sound: it needs air, water or something solid to travel through.
What turns sound into noise is context. Rain on a window is soothing at midnight and infuriating during a phone call. A neighbour's bass line is music to them and a problem for you. For public health purposes there is a harder line — sound loud enough to damage hearing, or persistent enough to disrupt sleep and raise stress responses, counts as noise whether anyone objects to it or not.
Indoors at night, the recurring offenders are predictable:
- Traffic and transport — road, rail and aircraft, the three sources the WHO models separately
- Neighbours and housemates — footsteps, doors, televisions, plumbing, different schedules
- The building itself — boilers, fridges, extractor fans, pipes expanding as they cool
- Your own household — a snoring partner, a restless pet, a phone that was not silenced
How is noise measured?
Noise is measured as sound pressure level (SPL), expressed in decibels. The decibel is not a unit of loudness in the way a metre is a unit of length — it is a ratio, on a logarithmic scale, comparing the pressure of a sound against a fixed reference point.
The reason for the logarithm is range. The quietest sound a healthy young ear can detect is a pressure fluctuation of about 0.00002 pascals. The threshold of pain is around 20 Pa, and a jet engine at close range produces more still. That is a spread of roughly a million to one. Worse, human hearing does not respond in a straight line: doubling the pressure does not feel twice as loud. A logarithmic scale compresses that unwieldy range into a usable 0–140 and tracks perception far better.
The formula is:
SPL = 20 × log10(p / pref) dB
where p is the root-mean-square sound pressure in pascals and pref is the reference pressure, 0.00002 Pa (2 × 10−5 N/m²) — the nominal threshold of human hearing. Because it is a ratio against that fixed point, 0 dB does not mean "no sound". It means "sound at the quietest level we can typically hear".
| Sound pressure level (dB) | Sound pressure (Pa) | Roughly equivalent to |
|---|---|---|
| 0 | 0.00002 | Threshold of hearing |
| 20 | 0.0002 | A very quiet rural bedroom |
| 40 | 0.002 | A library, a fridge humming |
| 60 | 0.02 | Conversation at arm's length |
| 80 | 0.2 | Kerbside traffic |
| 100 | 2 | A motorbike accelerating past |
| 120 | 20 | Threshold of pain |
| 140 | 200 | Jet engine at close range |
What does dB(A) mean?
You will usually see noise quoted as dB(A) rather than plain dB. The A-weighting filter discounts very low and very high frequencies, because the ear is genuinely less sensitive to them, and leaves the mid-range where speech sits. It is the weighting used in UK noise-at-work law and in almost every published noise limit, so unless a figure says otherwise, assume it is A-weighted.
Decibel chart: how loud is everyday life?
Below are typical A-weighted levels for sounds you are likely to meet. Treat them as indicative — measured levels vary enormously with distance, room acoustics and the specific machine involved.
| Level | Sound | What it means for sleep |
|---|---|---|
| 10 dB | Normal breathing | Below most people's waking threshold |
| 20 dB | Rustling leaves, a ticking watch | Genuinely quiet |
| 30 dB | A whisper, a quiet suburban bedroom | The WHO bedroom guideline1 |
| 40 dB | Fridge hum, light rain | Already above guideline; arousals become measurable |
| 50 dB | Quiet office, moderate rainfall | Disruptive for light sleepers |
| 60 dB | Conversation, dishwasher | Will wake most people |
| 70 dB | Vacuum cleaner, busy road through a closed window | Sleep is not realistic unattenuated |
| 80 dB | Kerbside traffic, alarm clock | Lower action value at work3 |
| 85 dB | Heavy traffic, food blender | Upper action value; hearing protection required at work3 |
| 90 dB | Lawnmower, lorry passing close | Prolonged exposure damages hearing |
| 100 dB | Nightclub, motorbike | Minutes, not hours |
| 110 dB | Chainsaw, live music at the front | Damage in under two minutes unprotected |
| 120 dB | Siren at close range, thunderclap | Threshold of pain |
| 140 dB | Firework at close range, jet engine | Immediate risk of permanent damage |
How loud is too loud?
There are two separate answers, because hearing damage and sleep disruption happen at very different levels.
For your hearing
In the UK, the Control of Noise at Work Regulations 2005 set the thresholds employers must act on.3 They are a useful yardstick outside work too:
- 80 dB(A) daily or weekly exposure — lower action value; hearing protection must be made available
- 85 dB(A) — upper action value; protection becomes mandatory and zones must be marked
- 87 dB(A) — exposure limit value, measured at the ear behind any protection worn
Peak pressures have their own limits: 135, 137 and 140 dB(C) respectively. Note that exposure is a dose, not a level — it combines how loud with how long. Eight hours at 85 dB and four hours at 88 dB carry roughly the same risk.
Three rough field tests, none of which requires a meter:
- You have to raise your voice to be understood by someone a metre or so away
- Your ears ring, feel full, or ache after you leave
- Sound seems muffled for a few hours afterwards — a temporary threshold shift, and a warning
For your sleep
Far lower. The WHO's 2018 guidelines for the European Region recommend keeping night-time road traffic noise below 45 dB Lnight, with lower figures again for rail and aircraft.2 Inside the bedroom, the long-standing guideline is around 30 dB for continuous background sound, with individual events kept under 45 dB.1 Sudden noises matter more than steady ones: a single door slam at 55 dB will do more damage to a night's sleep than a fan running at 45.
What does noise actually do to your sleep?
It fragments it. Noise triggers brief arousals and shifts you into lighter stages of sleep, and it does this at levels well below the point where you would wake up properly or remember anything in the morning.5 You can lose a meaningful amount of deep sleep to a road you have stopped consciously noticing.
The next day
The changes you cannot feel produce effects you can. Studies of people sleeping under traffic and aircraft flight paths consistently report next-day sleepiness, low mood, irritability and slower performance.67 The structure of the night was disturbed even where the sleeper's own account of it was not.
Over years
Here the picture needs care. Chronically poor sleep — from any cause — is associated with raised blood pressure, cardiovascular disease, weight gain and type 2 diabetes.8 The evidence linking noise specifically to those outcomes is thinner and largely observational, but it points the same way: the HYENA study found higher rates of hypertension among residents near airports,9 and later multi-country work reported associations between aircraft and road noise and cardiovascular disease.1011 These are correlations, not proof of cause — but they are consistent enough to take seriously.
"A quiet space is an essential component of an environment that is conducive to sleep. I always try to get a detailed description of my patient's sleep environment so we can identify any factors that might be creating disruptive noise."
Is it better to sleep in silence or with sound?
Silence wins in principle and often loses in practice. A truly quiet bedroom is the ideal, but very few people have one, and a room that is quiet on average can still be full of the sudden, irregular sounds that wake you. Adding a steady background sound — masking — raises the noise floor so those spikes stand out less against it. It also gives an anxious mind something to settle on.
The research is genuinely mixed, and it is worth knowing what each study found rather than taking "white noise works" on trust:
- A home-based study of 48 healthy young adults found that 43 dB air-conditioner noise produced no significant improvement in how quickly they fell asleep, how long they slept, or sleep efficiency.12 Absence of a detectable benefit, note — not evidence of harm.
- In a small coronary care unit study (n = 60, not randomised), sleep quality scores worsened sharply over three nights in the usual-care group while staying essentially flat in the group given an hour of white noise.13 The honest reading is that masking blunted a decline rather than improved sleep outright.
- In a crossover trial, 18 healthy adults put to bed 90 minutes early — a laboratory model of short-term insomnia — fell asleep a median 38% faster with broadband sound playing, dropping from 19 minutes to 13.14 A real effect, but on healthy sleepers in an artificial set-up, and about six minutes in absolute terms.
So: no guarantees, small samples, and effects that depend on who you are. Masking is cheap and reversible, which makes it worth a fortnight's trial — but if it does not suit you, the evidence gives you no reason to persevere. Blocking the sound is the more dependable route, and the two are not mutually exclusive.
What are the best sounds to sleep to?
Whatever you choose, the same two rules apply: it needs to be loud enough to cover the sounds you are hiding, and steady enough that it never itself becomes an event. Anything with sudden changes in volume — a playlist with a loud track four songs in, a podcast, a radio station with adverts — defeats the purpose.
White noise and pink noise
White noise contains every audible frequency at roughly equal energy, which is what gives it that flat hiss. Plenty of people find it harsh. Pink noise weights the balance towards the lower frequencies, producing a softer, deeper sound closer to steady rain or distant surf, and it suits most sleepers better as a first try.
A fan or air conditioner
The oldest masking device there is, and it does produce a genuinely constant sound. Bear in mind the air-conditioner study above found no measurable sleep benefit — the appeal may be as much thermal as acoustic.12
Music
Highly personal, and the usual failure is arrangement rather than taste: tracks with dynamic range pull you back towards wakefulness. If you use music, pick something with a narrow volume range and set a sleep timer.
Natural sound
Rain, waves and wind work well for the same reason pink noise does — broad frequency content with slow, gentle variation. Avoid recordings with birdsong or thunder, which are made of exactly the sudden events you are trying to mask.
How to block out noise at night
Work outward from the ear. Every method below either absorbs sound in the room, stops it entering the room, or stops it entering your ear — and the last of those is the only one that works regardless of where the noise is coming from.
Soften the room
Hard surfaces reflect sound and keep it alive in a room; soft ones absorb it. Rugs on bare floors, upholstered furniture and heavy curtains all reduce reverberation, which makes a room feel calmer even when the noise source is unchanged. This helps most with sound generated inside your home. It will do very little against a lorry outside.
Seal the gaps
Outside noise mostly arrives through openings, not walls. Draught-proofing strips around a window frame and a brush seal under the bedroom door are cheap and often deliver more than people expect. Secondary glazing is the serious version — an independent inner pane with an air gap, considerably more effective against traffic than swapping one double-glazed unit for another.
Silence the household
A fridge or extractor fan that has started making a new sound is usually failing, not settling; get it looked at. When you replace an appliance, check its declared noise output, which is now on most spec sheets. And put every device in the bedroom on silent — not vibrate, which is louder against a hard bedside table than most ringtones.
Agree the rules
Where the noise is another person, the fix is a conversation rather than a purchase. Agreed quiet hours, headphones for late television, and preparing tomorrow's clothes and meals before bedtime all remove predictable disruption. If a partner's snoring is the problem — particularly if it comes with choking or gasping — that warrants a GP visit rather than a workaround, as it can indicate obstructive sleep apnoea.
Block it at the ear
Earplugs are the most reliable option available to most people, because they do not depend on owning the building or persuading anyone. A randomised trial in intensive care gave 64 patients earplugs and an eye mask: those patients had significantly fewer prolonged awakenings, and while the trial's headline deep-sleep measure did not reach statistical significance, the patients who kept their earplugs in all night did record more deep sleep.15 That last detail is the practical one — attenuation you cannot tolerate is attenuation you do not get.
"For those struggling with a noisy sleep environment, be prepared for some trial and error. For example, you might find ear plugs to be helpful at first, but over time you may find you prefer the noise of a fan. Go into this with an open mind and you can find the right solution, because you deserve a restful and quiet sleep environment."
Hearing protection ratings, explained
Attenuation figures on packaging are the most misread numbers in this category. Here is what they mean.
What is an NRR?
NRR stands for Noise Reduction Rating. It is the American single-figure rating for hearing protectors, required on packaging by the US Environmental Protection Agency and derived from laboratory testing. A higher number means more attenuation under test conditions.
A common misreading is worth correcting: an NRR of 29 does not mean sound below 29 dB is eliminated and everything above it gets through. Attenuation is subtractive, not a cut-off. The plug reduces what arrives at your eardrum across the spectrum — unevenly, since every protector performs differently at different frequencies.
What should I look for in the UK?
SNR, the Single Number Rating, is the European and UK equivalent, measured to EN ISO 4869-2 and carried by protectors certified to the EN 352 series. If you are buying hearing protection in Britain, SNR — with the accompanying H, M and L values for high, medium and low frequencies — is the figure that applies. NRR is useful mainly for comparing against American products.
Will I get the rated attenuation in real life?
No, and this is the part manufacturers rarely lead with. Laboratory ratings come from carefully fitted protectors on trained subjects. HSE guidance instructs employers to assume the level at the ear will be 4 dB higher than the manufacturer's data predicts, precisely to account for the gap between laboratory and real-world fit.4 Fit matters more than the number on the box.
Do earmuffs protect better than earplugs?
Not inherently. A well-fitted earplug can match or beat a muff across most of the frequency range. The meaningful differences are practical: muffs are easier to fit correctly and quicker to take on and off, plugs perform better at low frequencies and are the only realistic option if you sleep on your side. Compare the specific products, not the categories.
Is anything on the market that blocks 100% of noise?
No. The highest NRR available from a foam earplug sits at around 33 dB, and no wearable product blocks sound entirely. Part of the reason is that your body conducts sound to the inner ear through bone and tissue, bypassing the ear canal altogether — which sets a hard ceiling on what anything worn in or over the ear can achieve. The goal is bringing noise down to a level your sleeping brain ignores, not silence.
Can my ears get used to noise?
No. If a noise seems less loud after prolonged exposure, that is your hearing temporarily dulling, not toughening. Ears do not adapt to loud sound in any protective sense, and a temporary threshold shift is an early warning rather than a sign of resilience.
I have already lost some hearing — is protection still worth it?
Yes. Existing hearing loss confers no immunity against further damage. Noise-induced loss is permanent and no protector reverses it, but protection prevents additional damage and gives temporary threshold shifts a chance to recover before they become permanent.
References
- World Health Organization. (1999). Guidelines for Community Noise. Guideline values for bedrooms: 30 dB LAeq for continuous background noise, 45 dB LAmax for individual events.
- World Health Organization Regional Office for Europe. (2018). Environmental Noise Guidelines for the European Region. Read more.
- The Control of Noise at Work Regulations 2005, SI 2005/1643, reg. 4. Read more.
- Health and Safety Executive. (2021). Controlling noise at work (L108), 3rd ed. See also HSE, "Accounting for 'real world' factors". Read more.
- Halperin, D. (2014). Environmental noise and sleep disturbances: A threat to health? Sleep Science, 7(4), 209–212. Read more.
- Basner, M., Müller, U., & Elmenhorst, E. M. (2011). Single and combined effects of air, road, and rail traffic noise on sleep and recuperation. Sleep, 34(1), 11–23. Read more.
- Basner, M., Clark, C., Hansell, A., Hileman, J. I., Janssen, S., Shepherd, K., & Sparrow, V. (2017). Aviation noise impacts: State of the science. Noise & Health, 19(87), 41–50. Read more.
- Medic, G., Wille, M., & Hemels, M. E. (2017). Short- and long-term health consequences of sleep disruption. Nature and Science of Sleep, 9, 151–161. Read more.
- Jarup, L., Babisch, W., Houthuijs, D., Pershagen, G., Katsouyanni, K., Cadum, E., … HYENA study team. (2008). Hypertension and exposure to noise near airports: The HYENA study. Environmental Health Perspectives, 116(3), 329–333. Read more.
- Floud, S., Blangiardo, M., Clark, C., de Hoogh, K., Babisch, W., Houthuijs, D., … Hansell, A. L. (2013). Exposure to aircraft and road traffic noise and associations with heart disease and stroke in six European countries: A cross-sectional study. Environmental Health, 12, 89. Read more.
- Franssen, E. A., van Wiechen, C. M., Nagelkerke, N. J., & Lebret, E. (2004). Aircraft noise around a large international airport and its impact on general health and medication use. Occupational and Environmental Medicine, 61(5), 405–413. Read more.
- Alkahtani, M. N., Alshathri, N. A., Aldraiweesh, N. A., Aljurf, L. M., Aldaej, L., Olaish, A. H., … BaHammam, A. S. (2019). The effect of air conditioner sound on sleep latency, duration, and efficiency in young adults. Annals of Thoracic Medicine, 14(1), 69–74. Read more.
- Farokhnezhad Afshar, P., Bahramnezhad, F., Asgari, P., & Shiri, M. (2016). Effect of white noise on sleep in patients admitted to a coronary care unit. Journal of Caring Sciences, 5(2), 103–109. Read more.
- Messineo, L., Taranto-Montemurro, L., Sands, S. A., Oliveira Marques, M. D., Azarbarzin, A., & Wellman, D. A. (2017). Broadband sound administration improves sleep onset latency in healthy subjects in a model of transient insomnia. Frontiers in Neurology, 8, 718. Read more.
- Demoule, A., Carreira, S., Lavault, S., Pallanca, O., Morawiec, E., Mayaux, J., Arnulf, I., & Similowski, T. (2017). Impact of earplugs and eye mask on sleep in critically ill patients: A prospective randomized study. Critical Care, 21(1), 284. Read more.