ToneToolkit

Decibel Meter

Estimate how loud a room is, with your own calibration offset.

Everything runs in your browser. No audio is ever uploaded.

This is an estimate, not a measurement. Browser microphones are not calibrated, every device has a different sensitivity, and the operating system may apply automatic gain control or noise suppression that moves the numbers on its own. Use a real sound level meter for anything that matters, such as workplace exposure.

Measure the sound level

Allow microphone access so the meter can listen. The audio is measured on your device and is never recorded or uploaded.

Capture stops as soon as you leave or close this page.

A decibel meter turns your microphone into a rough sound level readout. This online sound level meter, also called a dB meter or a noise level meter, shows the level around you in decibels, keeps the minimum, average and maximum since you started, and lets you set your own calibration offset.

How to use it

  1. Press Allow microphone and accept the browser prompt. Everything is measured on your device, and nothing is recorded or uploaded.
  2. Hold the device still, away from your hands and clothing, and point the microphone toward the sound. Covering the microphone with a finger is the single most common reason a reading looks wrong.
  3. Read the large number and the bar under it. Minimum, average and maximum accumulate from the moment you start, and the reset button clears all three when you move to a new spot.
  4. Set the calibration offset. The default of +90 dB puts normal conversation near 60 dB on most laptops and phones. If you have a real sound level meter, put both in the same place and move the offset until the numbers agree.
  5. Check the processing note under the controls. If your browser reports automatic gain control or noise suppression as active, it is quietly raising soft sound and holding loud sound back, and the readings follow that as much as the room.

What you can use it for

Comparing two rooms or two times of day. Absolute accuracy hardly matters when you want to know whether the back bedroom is quieter than the front one, because the same device and offset make the comparison fair.

Checking a noisy appliance. Run the meter beside a fridge, a fan or a computer, note the average, then change something and compare. The difference is more reliable than either number on its own.

Getting a sense of your listening level. Play music at your usual volume and watch the average. Many people are surprised by how high it sits once they see a number.

Documenting a complaint informally. A screenshot of a maximum is not evidence, but it does tell you whether a proper measurement is worth arranging.

Teaching what decibels mean. Clap, whisper, run a tap and watch the scale move. The logarithmic scale becomes much easier to grasp when you can see it react.

What the numbers mean

LevelWhat that sounds like
30 dBA quiet bedroom or a library.
60 dBNormal conversation about a meter away.
70 dBA vacuum cleaner or a busy restaurant. Loud enough that people raise their voices, but not itself a risk.
85 dBBusy traffic. Sustained exposure at this level risks hearing damage over a working day.
100 dBA concert or power tools. Even short exposure can damage hearing.
120 dBA siren close by or the front row of a loud concert. Around the point where sound becomes physically painful.

Accuracy and limits

This is an estimate, not a measurement. A browser microphone has no calibration, no known sensitivity and no standard frequency weighting, so the figure depends on your hardware and on the offset you chose. Two phones side by side can differ by more than 10 dB.

Browser processing moves the numbers on its own. Automatic gain control raises quiet passages and holds loud ones back, and noise suppression removes steady background sound, which is exactly the thing you were trying to measure. The note on the page reports what your browser admits to.

Anything that matters needs a real instrument. Workplace noise assessments, legal noise complaints and hearing conservation programmes all use a calibrated sound level meter with A weighting and a defined time response, and none of that exists here.

Frequently asked questions

How accurate is a phone decibel meter?

Roughly, in the sense that it will tell a quiet room from a loud one. Without calibration the absolute figure can easily be 10 dB out in either direction. It is useful for comparisons on one device and unsuitable for anything official.

How do I calibrate it?

Put a sound level meter you trust in the same place, play a steady sound, and move the offset until this page shows the same figure. Without a reference, leave the offset at +90 dB, which puts conversation near 60 dB on typical hardware.

What is a safe noise level?

Standard occupational guidance treats 85 dB sustained across a working day as the point where hearing protection and exposure limits come into play, and every 3 dB above that roughly halves the time considered acceptable. That is general guidance rather than advice about you; a doctor or an audiologist can speak to your own situation.

Why does the reading move when nothing changes?

Partly because rooms are never truly still, and partly because automatic gain control adjusts continuously. Watch the average over several seconds rather than the instantaneous number, and reset the statistics whenever you move.

What is the difference between dB and dBFS?

dBFS measures a digital signal against full scale, so it is always zero or negative and says nothing about the room. The dB figure here is that value plus your calibration offset, which is an attempt to express it as sound pressure level in the room. The offset is the whole difference.

Can I use this for a noise complaint?

As a first look, yes; as evidence, no. An uncalibrated browser reading will not stand up anywhere. Use it to decide whether the problem is real and persistent, then ask your landlord, your employer or the local authority for a proper measurement.

Related tools