Can your phone measure decibels at all?
Yes. The microphone turns sound pressure into an electrical signal, and the app calculates the sound level in dB. The catch is elsewhere: phone microphones are optimised for speech, not for measurement. They aren't calibrated, they differ between models, and the operating system processes the signal before an app ever sees it.
How well it can still work was studied by NIOSH, the US National Institute for Occupational Safety and Health. In 2014, Kardous and Shaw tested sound meter apps in a reverberant chamber with noise between 65 and 95 dB against a reference microphone. Of more than 130 iOS apps reviewed, 10 met the selection criteria, and some of them were within ±2 dB(A) of the reference on average. On Android the authors found large differences between devices – because of the many different microphones and audio components.
The biggest source of error: automatic gain control
For calls and voice recordings, phones adjust the level automatically (AGC, automatic gain control): quiet voices are boosted, loud ones turned down. For a measurement that's fatal – the app sees almost the same signal at 60 and at 80 dB. On top of that, noise suppression filters out steady noise. But steady noise is exactly what road noise is.
How dB Ride avoids it
- Android: dB Ride uses the unprocessed microphone source when the phone offers it, otherwise a source with as little processing as possible. Automatic gain control and noise suppression are also switched off where the device allows it.
- iOS: the audio session runs in measurement mode, which bypasses the system's voice processing.
- Real A and C weighting: frequency weighting is applied as a filter to the signal according to IEC 61672, not as a flat correction.
- Clipping: when the signal reaches the microphone's limit, the app shows a warning.
Where the limits are
- Frequency response: phone microphones capture very low and very high frequencies less accurately, so low drone can be underestimated.
- Very high levels: from about 90–100 dB some microphones clip. Read values in this range – for example under a motorcycle helmet – with caution.
- Wind on the microphone: airflow directly at the mic creates noise of its own. Windows up, vents off, keep the phone out of the airflow.
- Case and position: a case covering the microphone or a different spot can change the result by several dB.
How to calibrate
Without calibration, absolute values can be off by several dB depending on the device. A reference lets you correct that:
Get a reference
Ideally a calibrated sound level meter. On iPhone, an alternative is the free NIOSH SLM app, which NIOSH tested in its own lab.
Measure at the same time
Both devices side by side, microphones pointing the same way, a steady sound such as noise or a fan, both in dB(A).
Set the offset
Enter the difference as the calibration offset in dB Ride's settings – up to ±15 dB in 0.5 dB steps.
What the readings are good for – and what not
A phone is strongest at comparisons with the same device in the same position: old vs. new tires, before vs. after sound deadening, 60 vs. 80 mph. Device-specific deviations then largely cancel out. The decibel chart helps you put values into context. For official measurements, expert reports or evidence for authorities, an app is not suitable – that requires calibrated instruments and standardised procedures.
dB Ride is not a certified sound level meter and doesn't claim to be. It's built to measure road noise reproducibly and compare it fairly – with GPS speed for every reading.
FAQ
How accurate is a phone decibel meter?
Without calibration, absolute readings can be off by several dB depending on the device. In a 2014 NIOSH study, the best iOS apps were within ±2 dB(A) of a reference meter in the lab. For comparisons with the same phone, readings are far more meaningful than as absolute values.
Why do different apps show different values?
Because microphone source, signal processing, frequency weighting and calibration differ. If an app uses the system's automatic gain control, level differences are heavily compressed.
How do I calibrate dB Ride?
Measure at the same time as a reference, such as a calibrated sound level meter or the NIOSH SLM app on iPhone, and enter the difference as the calibration offset. The range is ±15 dB in 0.5 dB steps.
Can phone measurements be used as legal evidence?
No. Official measurements and expert reports require calibrated instruments and standardised procedures. Phone readings are for orientation and your own before/after comparisons.
Measure it yourself – with honest numbers.
dB Ride is free for Android and iOS. Measuring, noise by speed, the map and comparing two recordings are free; Premium (one-time purchase, no subscription) removes ads and unlocks export and comparisons of up to 8 recordings.
Sources & context
- Kardous, C. A.; Shaw, P. B. (2014): Evaluation of smartphone sound measurement applications. J. Acoust. Soc. Am. 135(4), EL186–EL192 – doi.org/10.1121/1.4865269, free full text on PubMed Central
- NIOSH: New NIOSH Sound Level Meter App (iOS) – cdc.gov
- IEC 61672-1: Electroacoustics – Sound level meters (definition of A and C frequency weightings).
- Technical details of dB Ride (microphone source, filters, calibration offset): eastfreak Studios.