Decibel Sound Meter: questions and honest answers

Close enough to compare one room with another, follow a level going up and down and see when something is clearly loud. Not close enough for a legal, workplace or hearing-test measurement. Phone microphones are tuned for voices, vary between models and squash very loud bangs, so treat the number as an estimate.

Yes, by entering a correction rather than just noting it. Stand next to a calibrated meter in a steady noise and set the difference in Exposure > Calibration, which offers -15 to +15 dB in 0.1 dB steps and applies to every reading. The Meter tab also has Calibrate with Auto Calibrate for a quiet room and a -20 to +20 dB Manual Adjustment.

Because sound itself does — a voice, a door, a passing car, all changing the level from one moment to the next, and the meter follows with a Fast (125 ms) response. For one number that describes a period, record it and read AVG: an energy average (LAeq), not a simple mean, so short loud moments count for what they really are.

In the app's reference table, a Quiet Library is 40 dB, Normal Conversation 60 dB, a Vacuum Cleaner 70 dB, a Busy Street / Alarm Clock 80 dB, Heavy Traffic 85 dB, a Subway Train 100 dB and Power Tools / Chainsaw 110 dB. From 85 dB, the time you can safely spend there each day starts to shrink quickly.

Measuring works offline, and the meter analyses sound in memory without saving audio. The free version loads its ads over the internet, so an export you get by watching an ad needs a connection too; so do App Store purchases and the place name shown for a location. The Video tab does record video with sound, which stays on your phone unless you share it.

Treat it as your own organised log, not as proof. It records the date, duration, average and peak of each episode, with notes, location and optional video showing the dB level, which helps you describe the problem clearly. An authority or court may require a measurement by a professional with a calibrated meter, so ask what they accept.

No. The Noise app is Apple's own app on Apple Watch. Decibel Sound Meter runs on iPhone and iPad and has no Apple Watch version, so you measure with the phone or tablet itself.

Apps use different weightings, response times, averages and built-in corrections. This one uses A-weighting and Fast response and adds +10 dB on iPhone (+8 dB on iPad) because phone microphones read low. The only fair comparison is against a calibrated meter, and you can enter that difference as a calibration offset.

No. It applies the textbook NIOSH and OSHA dose formulas, but the input is a phone microphone, so the result is a screening estimate. The app labels it Screening tool, not a dosimeter. For a formal workplace noise assessment you need a worn dosimeter and a qualified person.

The camera is only for the Video tab. Location is optional and only tags where a recording was made and places it on the noise map. Tracking permission only affects how ads are personalised in the free version; saying no changes nothing in the meter.

No. It listens to the sound in the room through the phone's microphone and cannot see the level inside headphones or earbuds.

The meter shows -- when it is stopped, or when the system has taken the microphone, for example during a phone call. In that case you see Microphone paused by the system. Measuring resumes on its own. and the reading comes back by itself.

The last 100. When you save the 101st, the oldest one is deleted automatically. Exposure sessions are kept separately in Sessions until you delete them, so export anything you need to keep for longer.

Open a session in History and use Export as PDF, Export as CSV or Share as Image Card; each needs Premium or one short ad per export. Share Summary, a plain-text summary, is always free, and so is the CSV export of exposure sessions.

A running recording on the Meter tab and a running Exposure measurement both continue when you switch apps. Being warned while you are in another app is different: that needs Premium's Background Monitoring together with Noise Alerts switched on.

Only dBA. The app uses A-weighting, which follows how human hearing responds, with Fast time weighting. It has no C-weighting, no flat (Z) weighting and no frequency analysis, so bass-heavy noise such as a subwoofer reads lower than a C-weighted meter would show.

Not reliably. Very loud, very short sounds overload a phone microphone and read low, and the display is limited to 140 dB. Use the app for steady or longer noise, and a dedicated meter for impulse noise.