Snoring Apps That Flag Breathing Pauses and Their Limits

At a glance

What obstructive sleep apnea is, in one paragraph

The National Heart, Lung, and Blood Institute describes sleep apnea as a common condition in which your breathing stops and restarts many times while you sleep. In the obstructive type, your upper airway gets blocked many times during the night. In the central type, your brain doesn't send the signals needed to breathe. NHLBI lists breathing that starts and stops, frequent loud snoring and gasping for air as signs during sleep, and daytime sleepiness, dry mouth and headache as signs while you're awake. Left untreated, it raises the risk of stroke, heart attack and other serious problems, and a provider may order a sleep study to find out which type you have and how serious it is [1][2]. Snoring by itself doesn't mean you have it. Plenty of people snore with no pauses at all.

What a microphone can and cannot hear

A phone app works from one signal, the sound in the room, and it can hear breathing get louder, go quiet and come back with a snort. What it can't tell is why the room went quiet.

Silence isn't the same as no airflow

A sleep study measures breathing itself. The Sleep Foundation describes polysomnography as monitoring brain waves, blood oxygen, heart rate and breathing through the night [3], and a microphone measures none of those. When your breath goes silent, all the app sees is a gap in the audio. Maybe that's a pause in breathing. Or quiet breathing. Or your head turned away from the phone, or a blanket over your face.

Some events make no sound at all. Central apneas involve no breathing effort, so there's no struggle to hear, and partial blockages (the hypopneas that count toward a sleep study score) can reduce airflow without a clean silence. An audio app misses both.

Position and noise set the limits

• Distance. Breathing fades fast. When you want breathing pauses, Snore Timeline asks for the phone 1 to 2 feet from your head, closer than the 3 to 4 feet it suggests for snoring.
• Your position. Roll away from the phone and your breathing can drop below what the mic picks up. That looks like silence.
• Room noise. A fan, white noise machine, air purifier or HVAC can cover breathing. It can cover the silence, too. Snore Timeline's guide notes the app may miss quiet events once a room sits above 45 dB at rest.
• A bed partner. A second sleeper's breathing can fill your silence, so a real pause may go unflagged.

All of these push the count the same way. An audio app misses events more often than it invents them. So a low count isn't proof that nothing happened.

How Snore Timeline flags a breathing disruption

It takes three things, in this order, for Snore Timeline to log a breathing disruption:

1. Audible breathing for at least 1 minute. The app checks twelve 5-second windows and needs breathing in at least 11 of them. That proves your breathing was loud enough to hear before it went quiet.
2. Silence for 10 seconds or more. Shorter gaps never count. That rules out the normal pause between breaths.
3. A sound that ends the silence, like a loud snore, a gasp or a breath. A gasp, a snort or a sudden loud breath has to break the quiet, and that's what separates a pause from you rolling away from the phone.

Each disruption shows up as a gray-blue marker on your timeline. Tap one in the nightly summary to jump to that moment and listen. For every one, the app logs the time, how long the silence lasted and how strong the recovery breath was, and it grades each silence by length, with under 12 seconds counted as mild, 12 to 20 seconds as moderate and over 20 seconds as severe.

The summary also shows disruptions per hour of sleep, with its own bands from Minimal to Significant. That rate describes what the app heard. It isn't a medical score. It isn't an apnea-hypopnea index (AHI) either. The nightly assessment needs at least 2 hours of sleep, and if the app never hears enough breathing to form a baseline, it stays hidden. You'll find the full rules, including how CPAP nights and quiet breathers affect the count, on the Breathing Disruptions support page, and How Detection Works covers the decibel measurements.

I checked the flag against 57 overnight recordings from PSG-Audio, a public sleep-lab dataset with clinician-scored events [11]. Lab-scored events include hypopneas and quiet apneas that make no sound, so the flag catches only some of them. The app has not been clinically validated. Treat the count as a conservative estimate. Watch the trend across nights. If you pair an Apple Watch that records blood oxygen, the summary can show time below 95% saturation and your lowest reading, both of which come from the watch through Apple Health.

What other apps offer

I've described each product below from its own App Store listing or its maker's support page, and the purchases and privacy labels are developer self-declarations, as declared on the App Store on 1 November 2026. These listings and pages do not claim to diagnose sleep apnea.

SnoreLab

SnoreLab has 58K ratings on the US App Store and a long track record with its Snore Score and remedy tracking. Snore Timeline's answers to those are a 0 to 100 sleep score, where breathing disruptions cost more points than snoring, and remedy tracking with 23 built-in sleep factors plus any you create. It has far fewer ratings because it reached the App Store in September 2025, SnoreLab in 2012, and I build it alone with no outside funding and no marketing budget. SnoreLab's App Store listing describes BreathFlow as a breathing-stability measure that looks at sound patterns for signs of irregular breathing, plus a Breathing Rate feature that tracks your respiratory rate and picks out possible breathing interruptions. Both are part of SnoreLab Premium [4]. On its website, SnoreLab defines breathing disturbances as temporary interruptions of breathing patterns during sleep. The listing says SnoreLab is not a medical device and is not intended to diagnose or treat sleep apnea. For a side-by-side view, see Snore Timeline vs SnoreLab.

Somnio+

On the App Store, Somnio+ goes by "Sleep Apnea Detector - Somnio+". Its listing says it records sleep sounds and marks stretches of prolonged silence without breathing sounds as potential breathing pauses, and that it's not a medical device and does not provide a medical diagnosis [5]. Like Snore Timeline's, its privacy label reads Data Not Collected. It sells weekly and annual subscriptions from $4.99 to $29.99. Of the 14 snoring apps I checked, four declare Data Not Collected, and Snore Timeline is the only one of those four with no purchases.

Snollo

Snollo's App Store listing says it analyzes sound on the device and combines it with heart rate, respiratory rate and blood oxygen from an Apple Watch. Its version history lists a breathing indicator added to Sleep Sounds [12]. Its site says Snollo does not diagnose or treat any medical condition [13].

Apple Watch sleep apnea notifications

This one's built into watchOS. According to Apple Support, the watch uses its accelerometer to look for breathing disturbances while you sleep. Each night gets classified as Elevated or Not Elevated. Every 30 days the watch analyzes the data, and it notifies you if you have Elevated breathing disturbances on a consistent basis. The feature is meant for people 18 or older who haven't been diagnosed with sleep apnea, and Apple says it's not intended to diagnose, treat or aid in the management of sleep apnea [6]. Apple announced FDA 510(k) clearance for the notifications in September 2024 [7]. The watch doesn't record audio for this feature, so it complements a microphone. There's more on the watch in Does Apple Watch track snoring? and on the Apple Watch support page.

How accurate phone apps are

I haven't found a published study that tests any of the consumer apps above against a sleep study. The research tests lab algorithms running on phones. Results depend on the method.

Narayan and colleagues put an unmodified Samsung Galaxy S5 near the bed of 91 patients who were having a clinical sleep study. They built their sound algorithm on 32 patients, then tested it on 59 others. Compared with whole-night polysomnography, their abstract reports a c-statistic of 0.87 for separating an AHI under 15 from an AHI over 15. The authors conclude that phone-recorded sound can identify apnea and abnormal breathing. They call for future studies on screening [8].

Kim, Kim and Hwang pooled 11 studies of smartphone-based measurement (motion, sound, oximetry and combinations), and for moderate-to-severe obstructive sleep apnea (AHI above 15), their abstract reports pooled sensitivity of 0.9064 and specificity of 0.8801. The mean difference in AHI between phone and polysomnography was small. The limits of agreement were wide, though. The authors caution clinicians about making clinical decisions based on these devices [9].

Two things follow from that. The pooled figures mix methods, so they don't describe a microphone-only app. They also concern moderate-to-severe cases, with an AHI above 15. An app that flags silences can get a conversation started with your doctor. It can't give you the number a sleep study gives.

STOP-BANG and other screeners

STOP-Bang is a questionnaire. It began as the four-question STOP tool, which Chung and colleagues published in 2008 for surgical patients, covering Snoring, Tiredness during the day, Observed apnea, and high blood Pressure. In the same study, adding BMI, age, neck circumference and gender raised its sensitivity [10], which gives you eight yes-or-no items scored from 0 to 8.

A screener sorts people by risk. It doesn't measure your breathing. The 2008 study scored patients against the AHI from a monitored sleep study, and that sleep study is the test. If your score is high, or a partner has seen you stop breathing, the next step is a clinician and a sleep study. A low score with loud snoring and gasping is still worth bringing up.

What to bring to a doctor

A doctor gets more out of a pattern than out of one night. Bring:

• Several nights of data. Record a handful of full nights with the phone 1 to 2 feet away in a quiet room. The Doctor-Ready Data guide walks through setup.
• The flagged times. The Detected Events export lists every disruption with its timestamp, silence length and recovery strength.
• Exported clips. Hearing a silence end in a gasp tells a doctor more than a row of numbers does. Export & Sharing covers the CSV files, per-episode clips and full-night audio, all of which export at no cost.
• Partner observations. Note what a bed partner saw or heard, and on which nights. Add your own daytime sleepiness or morning headaches.

Be upfront about what the data is. It's phone audio that catches events loud enough to hear, and it tends to undercount. It all stays on your phone until you share it.

Frequently asked questions