How to Measure Decibels With Your Phone
A phone can be a useful sound meter when you use it carefully. This guide shows how to measure, calibrate, and interpret decibel readings without pretending a phone is a laboratory instrument.
Updated· 7min read
What a phone can and cannot do
If you want to know whether a room is quiet enough to sleep, a classroom is getting too loud, or a workshop is entering hearing-risk territory, a phone can give useful decibel readings. Open the online sound meter, allow microphone access, and you can see A/C/Z weighting, Fast or Slow response, Leq, Min, Max, Peak, a timeline, and an octave spectrum without installing a separate app.
The important word is useful. A phone is not the same as a class 1 or class 2 instrument built to IEC 61672. Its microphone was designed for speech, calls, and video, not certified acoustics. Still, if you use consistent technique and calibrate against a reference meter, it can turn vague impressions into repeatable measurements.
As a realistic rule: uncalibrated phones are commonly about ±3–7 dB in the 40–90 dB range. After per-microphone calibration, steady sounds may be within about ±1–2 dB. Below roughly 30 dB, microphone self-noise becomes a limit. Above roughly 110 dB, phone inputs may clip and under-report peaks. iOS also applies audio processing that cannot be fully disabled, so iPhone readings need extra caution.
Choose the right measurement settings
Before you worry about technique, set the meter for the job. Most everyday measurements should use A-weighting, because dB(A) follows the broad frequency sensitivity of human hearing and is widely used in environmental noise guidance, including WHO documents from 1999 and 2018. Use C-weighting when low-frequency energy matters, such as music bass or impulse peaks. Use Z-weighting when you want an unweighted view of the microphone signal.
Response speed matters too. Fast response reacts quickly and is useful when a sound changes from second to second. Slow response smooths the display and is easier to read for steady or slowly varying noise. Leq is often the most useful number because it averages acoustic energy over the measurement period. A dog bark may push Max high, while Leq tells you how much noise was present across the whole recording window.
| Setting | Good default | Why it helps |
|---|---|---|
| Weighting | A | Matches most environmental and comfort guidance |
| Response | Slow | Easier to read for rooms, traffic, neighbors, and appliances |
| Time | 1–10 minutes | Captures variation instead of one lucky second |
| Main metric | Leq | Represents average sound energy over the measurement |
| Extra metrics | Max and Peak | Shows brief loud events that the average can hide |
For more background on the scale itself, read what is a decibel. If you just want familiar reference points, the decibel chart and decibel levels pages are good companions.
Set up the phone like a small instrument
Measurement technique can change the result by several decibels. Put the phone at about ear height: usually 1.2 to 1.5 meters above the floor for seated or standing listening positions. Keep it at least 1 meter from walls, large furniture, and windows when possible. Hard surfaces reflect sound; measuring too close to them can make the level higher or lower depending on reflections.
Hold the phone still, or better, place it on a stable stand. Do not cover the microphone with your fingers, case flap, clothing, or bedding. Many phones have more than one microphone, and the active input may not be where you expect. If the number drops when you rotate or hold the phone differently, your hand or case may be changing the sound path.
Avoid measuring with the phone flat on a table unless that is the situation you specifically care about. A tabletop can reflect high frequencies into the microphone. If you must use a table, keep the setup the same each time so comparisons remain meaningful.
Outdoor measurements add wind. Even a light breeze can create rumble at the microphone and inflate low-frequency readings. Professional meters use windscreens for a reason. With a bare phone, treat windy outdoor results as rough.
Measure long enough for the question
The most common mistake is reading the first number that appears. Real sound varies. A refrigerator cycles, traffic comes in waves, people pause between sentences, and music has quiet and loud passages. Start the meter, let the reading settle, then run it long enough to represent the situation.
For a steady appliance, 30–60 seconds may be enough. For a bedroom at night, use at least 10 minutes if you want a fair picture of background noise and intermittent peaks. For a classroom or office, a full activity period is better than a short sample. For a gym, bar, or concert, take repeated readings at different locations.
Use the timeline to spot changes, then compare Leq, Max, and Peak. Leq answers, “What was the overall level?” Max answers, “How loud did it get on the normal time response?” Peak answers, “What was the highest instantaneous pressure the phone detected?” These are related but not interchangeable.
Calibrate step by step
Calibration is the biggest improvement you can make if you need credible phone readings. It does not turn a phone into a certified IEC 61672 sound level meter, but it can correct the microphone sensitivity offset for the device you are actually using.
Use a reference sound level meter if possible. Place the phone and reference meter side by side, with microphones close together but not touching. Put both at the same height, away from walls, and expose them to the same steady sound. A fan, pink-noise speaker, or stable room source is better than speech.
- Open the online sound meter and allow microphone access.
- Select the same weighting and response as the reference meter, usually dB(A) Slow.
- Wait until both readings are reasonably stable.
- Tap the Calibrate button.
- Type the reading shown on the reference meter.
- Choose “Match now” to set the offset automatically.
- If needed, use the slider for a small manual adjustment.
Calibration is stored per microphone. If you change from the built-in microphone to a USB microphone, headset, Bluetooth device, or another browser input, calibrate again. Also recalibrate after major operating-system changes or if the phone case changes the microphone opening.
iPhone, Android, and browser notes
Android devices vary widely because manufacturers choose different microphone parts and signal chains. Some models behave well in the speech range but struggle with bass or very loud sound. Calibration is especially helpful on Android because the uncalibrated offset can differ greatly by model.
iPhone hardware is more consistent across devices, but iOS audio processing is the special issue. The system may apply processing intended for communication, and web apps cannot fully disable all of it. That does not make iPhone readings useless. It means you should be cautious with very low levels, sudden peaks, and comparisons against Android phones.
In any browser, grant microphone permission only to sites you trust. This site analyzes audio in your browser; audio is never recorded or uploaded. You can install it as an offline app for field use, then use CSV export, snapshot images, and share features when you need to document a session.
Accuracy limits to write down
When you share a reading, include the limits. A good note is stronger than a bare number: “iPhone 14, A-weighted Slow, calibrated to a class 2 meter, 10-minute Leq 48 dB(A), Max 61 dB(A), bedroom center, windows closed.” That tells another person how to interpret it.
Remember the practical boundaries:
| Situation | What can go wrong |
|---|---|
| Below about 30 dB | Phone self-noise can dominate the reading |
| 40–90 dB | Best practical range for many phones |
| Above about 110 dB | Clipping may make loud sound read too low |
| Low-frequency bass | A-weighting may understate annoyance or vibration |
| Impulse noise | Phone processing may miss the true peak |
| Legal or workplace compliance | Certified instruments and defined methods may be required |
For hearing-risk contexts, compare your readings with established guidance such as the NIOSH Recommended Exposure Limit of 85 dBA for 8 hours with a 3 dB exchange rate, and OSHA’s 90 dBA Permissible Exposure Limit with a 5 dB exchange rate. For a deeper listening-safety guide, see safe decibel levels for listening.
A simple field checklist
Use the same checklist every time:
- Select A-weighting unless you have a reason to use C or Z.
- Use Slow for steady rooms and Fast for rapidly changing sound.
- Put the phone 1.2–1.5 m high and at least 1 m from walls when possible.
- Keep the microphone uncovered and the phone still.
- Measure long enough to capture the real pattern.
- Save Leq, Max, Peak, location, settings, and time.
- Calibrate against a reference meter when the number will influence a decision.
Used this way, a phone sound meter becomes more than a curiosity. It becomes a practical notebook for sound: not perfect, not certified, but consistent enough to help you understand what is happening and decide what to do next.