How to sleep with noise you can't turn off

Sound is not one-size-fits-all, and the clearest evidence for that comes from a study where it went both ways at once. Across 13 randomized studies, white and pink noise gave 335 young people with ADHD a small benefit on attention tasks; in the non-ADHD comparison groups of 11 of those studies, a similar number of participants took a small penalty (Nigg 2024). Same sound, opposite results. That is daytime task data rather than sleep data, and it is the cleanest demonstration anywhere that one sound can cut both ways. We make sound apps, and this is the page where we go through the four situations in which our own product is the wrong answer, or at least not the first one.

If your question is which sound to play rather than what to do about a sound you can’t escape, its counterpart is the better page: sleeping with sound covers the menu.

The noise is outside and it won’t stop

Traffic, neighbors, a road that gets loud at 6am. Here the mechanism is well understood: what wakes you is the size of the jump from background to peak, not the peak itself. Five sleepers in one lab study stirred at the same step up in level, 17.7 dB against 17.5 dB, whether or not white noise had been added to raise the background, and that added noise cut their arousals from 48 an hour to 16 (Stanchina 2005). Across pooled polysomnography of 94 sleepers and more than 25,000 noise events, the odds of waking rose about 35% for every 10 dB increase in indoor peak level (Basner & McGuire 2018).

Two levers follow, and most articles only give you one. You can lower the peak, or you can raise the floor. A seven-night study of 25 adults tested both: earplugs mitigated nearly all the effects of environmental noise on sleep and only began failing at the loudest exposure, around 65 dBA, while pink noise reduced REM sleep and made sleep structure worse when layered on top of the noise (Basner 2026). Twenty-five people is a small study, and it points the opposite way from how our category usually sells itself. Reach for attenuation first when the noise is loud. Choosing between a masking device and a blocking one is its own decision, and a one-night problem like fireworks or a party next door calls for blocking rather than masking almost every time.

The noise is a person, and they’re in your bed

A snorer is a different problem because the sound is loud, irregular, and close. The disruption is measurable: when ten couples were monitored on polysomnography simultaneously, treating the snoring partner’s apnea dropped the spouse’s arousal index from 21 to 12 an hour and lifted their sleep efficiency from 74% to 87% (Beninati 1999). Ten couples in 1999, so hold it loosely, and note the detail that survives anyway: spouses improved even though they had spent years habituating to the noise.

Here is where we will not sell you anything. No published study has tested whether a masking sound helps the partner of a snorer, and the mechanism above predicts it would struggle, because a snore peak is high and irregular and the masker would have to be loud to close that gap. What the evidence does support is treating the snoring. Loud habitual snoring together with witnessed pauses in breathing, gasping, or daytime sleepiness is the pattern that warrants a screening conversation with a doctor; the STOP-Bang questionnaire catches about 93% of moderate-to-severe sleep apnea, and about 84% of cases overall, while flagging plenty of people who don’t have it, so a positive result means get assessed rather than assume the worst (Chung 2008).

The noise is inside your own head

Tinnitus affects sleep for about half the people who have it, pooled across seven studies and 3,041 participants, though the studies disagreed enough that the pooled figure carries a confidence interval running from 40% to 67% (Gu 2022).

Sound therapy is the standard suggestion, and the standard suggestion has a hole in it. A Cochrane review of eight randomized trials covering 590 people found no evidence that sound therapy beats a waiting list, a placebo, or plain information, and noted that not one of the trials measured sleep quality at all (Sereda 2018). The 2024 US veterans’ guideline still suggests therapeutic sound for tinnitus self-care, at its weakest recommendation strength. Where that guideline does specify a level, in an audiologist-delivered protocol rather than in self-care, it is partial masking set to the lowest level that gives relief rather than something loud enough to drown the tinnitus out.

The nearest thing to trial evidence is cognitive behavioral therapy aimed at tinnitus distress: across four randomized trials it moved insomnia severity scores 3.28 points as a side effect, consistently across studies (Curtis 2021). Read that carefully. None of those trials set out to treat insomnia, their participants started below the clinical threshold for it, and 3.28 points is short of the change clinicians call meaningful. Sudden hearing loss, sudden pulsatile tinnitus, or tinnitus in one ear are reasons to see an audiologist rather than to shop for an app.

Sound itself is the trigger

For some people the problem is not volume but a particular sound: chewing, tapping, breathing. Misophonia was given a consensus clinical definition in 2022 by an expert panel, as reduced tolerance to specific sounds and the things associated with them (Swedo 2022). How common it is depends heavily on where you draw the line, which is why two nationally representative surveys landed at 4.6% and 18%.

The sleep evidence is one study. Among 102 young people aged 8 to 17 with clinically significant misophonia, about 30% had clinical levels of sleep problems by parent report, far above the general youth population and a little below the 43% seen in young people with anxiety disorders (Wagner 2026). Beyond that, a systematic review of misophonia treatments found one randomized trial, one open-label trial, and 31 case studies (Mattson 2023). The only sound-based treatment in there is tinnitus retraining therapy, in two uncontrolled case reports, and no study in the review measured a sleep outcome. We would like to tell you whether masking helps or hurts at night. Nobody knows, and if a sound is a trigger for you, that is a conversation for a clinician rather than a settings menu.

Sound just doesn’t work on you

This is the most common of the four and the least discussed, because no one selling audio has an incentive to raise it. The Nigg finding at the top is the cleanest version: same noise, benefit for one group, penalty for another. You can see it in schoolchildren too: with 51 pupils tested one at a time at school, background white noise improved recall for the ten the teachers had rated inattentive and worsened it for the other forty-one (Söderlund 2010).

Responsiveness varies for other sounds as well. ASMR produces measurable changes in heart rate and skin conductance, and it produces them only in people who experience the tingles at all (Poerio 2018). Two related posts sit here rather than under the menu, because both are really about which listener a sound works on: whether background sound helps you focus depends on the same individual split, and calming music for dogs turns out to have the same problem in a different species.

If you have tried steady sound at a sensible volume for a couple of weeks and it has done nothing, believe that. Quiet is a legitimate answer, and whether running sound all night carries a cost of its own is a fair question that the REM finding above leaves open. Infant sleep plays by stricter rules on volume and placement, which white noise for babies sets out.

Where that leaves you

Four problems, four different first moves: attenuate the loud ones, treat the snoring, take tinnitus to an audiologist, and stop pushing sound at a problem it has never been shown to touch. If your noise is ordinary and intermittent and you want something to lay over it, that is the case sound is good at, and Deep Sleep Sounds will do it for free with 35 sounds on the Free Forever plan. If it doesn’t work for you after a fair trial, we would rather you closed the app than kept paying for silence you could have had for nothing.

---

Sources: Nigg et al. 2024 · Stanchina et al. 2005 · Basner & McGuire 2018 · Basner et al. 2026 · Beninati et al. 1999 · Chung et al. 2008 · Gu et al. 2022 · Sereda et al. 2018 · Curtis et al. 2021 · Swedo et al. 2022 · Wagner et al. 2026 · Mattson et al. 2023 · Söderlund et al. 2010 · Poerio et al. 2018. This is general information, not medical advice — persistent insomnia is worth taking to a doctor.