Noise and Sleep: The Decibel Levels That Wake You Up

Your Ears Don't Have Eyelids

You can close your eyes. You can't close your ears.

That's the whole problem in one sentence. All night, your brain keeps listening. It's checking whether the sound outside is a passing truck or something that needs you awake. Most of the time it decides "nothing," and you never find out it happened.

But "you never find out" is not the same as "no cost." A lot of what noise does to sleep happens below the level you'd ever remember in the morning.

So let's put actual numbers on it. How loud is too loud, what the guidelines say, and which fixes hold up when someone tests them properly.

Key Facts

• WHO's bedroom target is under 30 decibels for good quality sleep, and under 40 dB at night outside bedrooms to avoid health effects (WHO Europe)
• Measurable body reactions start around 33 decibels — barely above a whisper (WHO systematic review, 2018)
• Every 10 decibels roughly multiplies your odds of a sleep-stage shift toward waking by about 1.35 (same review)
• Electronic beeps wake people more than human voices at the same volume (Annals of Internal Medicine, 2012)
• The evidence for white noise as a sleep aid is rated "very low" quality (Sleep Medicine Reviews, 2021)
• Earplugs beat pink noise in a head-to-head lab trial, adding about 21 minutes of deep and REM sleep on noisy nights (Sleep, 2026)

The Numbers

Decibels are confusing because the scale isn't linear. A jump of 10 decibels is a large jump, not a small one. Here's what the levels mean in a bedroom, using numbers from sources that measured them.

Two things stand out. First, the numbers that matter for sleep are quiet. We're talking about whisper-to-conversation range, not concerts. Second, the useful bedroom target — under 30 decibels — is lower than a lot of city bedrooms ever get.

Waking Up vs. Being Woken Up

Here's the distinction that changes how you should think about this.

Full awakenings are the ones you remember. Truck goes by, you sit up, you check the clock. Annoying, obvious, and honestly not the main problem.

Arousals are brief shifts toward wakefulness that show up on brain recordings and never reach your memory. Your sleep stage gets lighter for a few seconds. Your heart rate ticks up. Then you slide back down. In the morning you'd swear you slept straight through.

The 2018 systematic review that fed into WHO's guidelines found that maximum sound levels as low as 33 decibels trigger physiological reactions during sleep [^3]. That is quieter than most people's idea of "quiet."

And these micro-events add up. The same review estimated that for every 10-decibel increase in the peak indoor sound level of an event, the odds of shifting toward wake or the lightest sleep stage went up by roughly 1.35 times — and the number was almost identical for aircraft, road, and rail noise [^3]. A 2022 update focused on how people report their own sleep found even stronger relationships per 10 decibels, particularly for railway noise [^6].

The point: a night that felt fine can still have been chopped up. That's why "I sleep fine next to the highway" isn't the argument people think it is.

Why Your Brain Keeps Listening

Your auditory system stays partly online through the night, but it isn't a dumb pass-through. Your brain actively gates what gets in.

One elegant demonstration: researchers recorded people during a quiet night, counted their sleep spindles — short bursts of brain rhythm that show up in lighter non-REM sleep — and then ran them through a noisy night. People who produced more spindles tolerated the noise better [^7]. The gate is real, and some people's gate is stronger than others'.

Which is partly why noise sensitivity varies so much between people sleeping in the same room, and why "just get used to it" isn't advice you can give with a straight face.

Steady Noise Is Easier Than Sudden Noise

The brain isn't reacting to loudness alone. It's reacting to change.

A constant hum gives your auditory system nothing new to evaluate. A sudden bang, a phone chime, a door — those are exactly the kind of events a sleeping brain is built to check on.

The best evidence for this comes from a hospital study. Researchers played 14 common hospital sounds — voices, alarms, phones, ice machines, traffic, helicopters — at controlled levels from 40 to 70 decibels while 12 healthy volunteers slept in a lab. Electronic sounds were more arousing than other sounds, including human voices ¹. Same volume, different response. The sound's character mattered as much as its level.

The study also found that responses varied by sleep stage. Sounds during deep sleep produced fewer arousals than in lighter non-REM sleep, and sounds played during REM produced a larger, more sustained heart-rate response ¹. Your vulnerability to noise moves around across the night.

Does White Noise Actually Help?

This is where I'm going to annoy some people.

The idea makes sense: play steady broadband sound, and sudden noises don't stand out as much because they're buried in it. Masking is a real acoustic phenomenon.

The evidence that it improves sleep is weak. A 2021 systematic review pulled together 38 studies on continuous noise as a sleep aid and graded the quality of evidence as very low — the authors noted this sits awkwardly against how widely these machines are sold and used. Results ranged from "improves sleep" to "disrupts sleep," and the authors specifically warned that continuous noise may negatively affect sleep and hearing ².

Then in 2026 a proper head-to-head trial landed. Twenty-five healthy adults, seven nights each, in a lab, with full sleep recordings and a crossover design so everyone got every condition. They compared quiet control nights, recorded environmental noise, pink noise at two levels, earplugs, and combinations.

Pink noise on its own at 50 decibels made sleep worse, cutting REM sleep by about 18.6 minutes and adding about 20.1 minutes of light non-REM sleep compared with quiet control nights. Added on top of environmental noise, it produced only minor improvements in fragmentation and failed to protect deep sleep or REM ³.

Earplugs were the winner. On noisy nights, foam earplugs gave participants about 21 more minutes of deep sleep plus REM sleep combined, and on most measures earplug nights weren't statistically different from quiet control nights. The protection broke down at the loudest condition, 65 decibels ³.

So: earplugs have real, measured support. White noise machines have popularity. Those aren't the same thing. If a sound machine helps you personally, keep using it — I'm not going to argue with your own nights. Just don't assume it's doing something the research hasn't shown.

This Is a Public Health Problem, Not a Personal Failing

An estimated 139 million people live with chronic environmental noise above WHO guideline levels, and noise-related sleep disturbance accounts for more than 900,000 healthy life-years lost every year across Europe alone ⁴.

If your bedroom faces a road, that isn't a discipline problem. It's an exposure problem, and it responds to exposure fixes: heavier curtains, sealing gaps around windows, moving the bed away from the shared wall, earplugs. We put the rest of it in our bedroom setup guide, and noise interacts with the other room variables more than people expect — a room that's too warm makes you a lighter sleeper, and a lighter sleeper is a more noise-sensitive one.

Where Komori Fits

Komori is a contactless sleep and wellness device we're building. No camera, no wearable, nothing on your body. It isn't a medical device and it doesn't diagnose anything.

Room noise is one of the environmental signals it's being designed to log, alongside temperature, humidity, CO2, and light — plus sleep position, movement and restlessness, and bed-exit events. It is not designed to measure breathing, heart rate, or blood oxygen.

The reason environment logging is interesting here is timing. If your restlessness clusters at 11:40 PM every weeknight, and your room's noise level spikes at 11:38, that's a lead worth chasing. Maybe it's a neighbor's routine. Maybe it's a heating system. When it ships, the value isn't a verdict — it's giving you a timestamp so you can go find the cause yourself. We wrote more about what movement patterns can and can't tell you.

Noise is one of the few sleep problems where the fix is often cheap and physical. Try earplugs for a week. If you still wake up exhausted in a quiet room, the problem isn't the room, and that's the point to bring it to your doctor.

---

Footnotes

1. Buxton OM, et al. "Sleep Disruption due to Hospital Noises: A Prospective Evaluation." Annals of Internal Medicine, 2012. 12 participants, 14 sounds at 40–70 dBA. ↩ ↩²

2. Riedy SM, et al. "Noise as a sleep aid: A systematic review." Sleep Medicine Reviews, 2021. 38 articles; GRADE quality of evidence rated very low. ↩

3. Basner M, et al. "Efficacy of pink noise and earplugs for mitigating the effects of intermittent environmental noise exposure on sleep." Sleep, 2026. Randomized crossover laboratory study, 25 participants. ↩ ↩²

4. Garcia Molina G. "Protecting sleep in a noisy world." Sleep, 2026. Editorial. ↩