Watch live sound levels from your microphone — relative decibel readings with A/Z weighting and the Fast/Slow response of professional meters.Learn more ▾Show less ▴
Sound level questions rarely need a certificate — they need a fair comparison. Is the new fridge louder than the old one? Which room is quietest for recording? How much noise does the street add at night?
This meter answers those questions with live, relative decibel readings and the same A-weighting and Fast/Slow behavior professionals use, while being upfront that a browser cannot replace a calibrated instrument.
Readings are relative, not calibrated dB SPL. Phone and computer microphones are uncalibrated and every device hears differently, so no browser tool can measure absolute sound pressure reliably. Compare places, moments, and changes — and use a calibrated class 1 or class 2 meter (IEC 61672) for any regulatory or occupational measurement.
Current level
—dB(A)
Relative reading — not calibrated dB SPL
−800
Min
—
Max
—
Peak
—
Press “Start measuring” to begin. Your browser will ask for microphone permission.
Frequency weighting?A-weighting follows the IEC 61672 curve that approximates how human hearing responds at different frequencies — the standard choice for environmental noise, shown as dB(A). Z-weighting is flat and treats all frequencies equally, shown as dB(Z).
Time weighting?How quickly the reading follows changes, per IEC 61672: Fast uses a 125 ms time constant and tracks short events; Slow uses 1 second and gives a steadier average.
dB
Advertisement
About This Tool
This sound level meter turns your microphone into a live, relative decibel display. It shows the current level plus running minimum, maximum, and peak values, with the A and Z frequency weightings and the Fast and Slow time responses defined for professional meters in IEC 61672.
It is honest by design: because consumer microphones are uncalibrated, it measures change and comparison rather than claiming absolute dB SPL.
Press “Start measuring” and allow microphone access when your browser asks.
Choose A or Z weighting and Fast or Slow response, then watch the live level with Min, Max, and Peak.
Optionally set a calibration offset to match a reference meter, and press Reset to start fresh statistics.
How to Use
Press “Start measuring” and allow microphone access when your browser asks.
Choose A or Z weighting and Fast or Slow response, then watch the live level with Min, Max, and Peak.
Optionally set a calibration offset to match a reference meter, and press Reset to start fresh statistics.
Methodology
The meter asks the browser for raw microphone input with echo cancellation, noise suppression, and automatic gain control all turned off, so the signal stays as unprocessed as your device allows.
Levels are computed from the live waveform as decibels relative to full scale (dBFS), where 0 dB is the loudest signal your device can pass. A-weighting applies the IEC 61672 curve across the frequency spectrum; Fast (125 ms) and Slow (1 s) smoothing use the exponential time weighting of squared signal power that the same standard defines. An optional whole-decibel offset aligns the scale with a reference meter.
Treat every number as a comparison, not a certificate. A reading of −30 dB on its own says little; the 12 dB jump when the air conditioner starts is real information.
Higher (less negative) numbers mean louder sound at the microphone. Min and Max track the quietest and loudest smoothed levels since you started, and Peak follows the single loudest instant. NIOSH research found that only a few smartphone apps came within ±2 dB of professional meters even in laboratory tests — so compare positions, moments, and changes, and leave certified measurements to calibrated instruments.
Find the quiet corner: measure 20 seconds at each candidate desk with Slow response and A-weighting. If one corner sits at −48 dB while the other reads −41 dB, the first is about 7 dB quieter — a clear, repeatable difference on the same device.
Document a noisy moment: with Fast response, watch Max while the neighbor's music plays and again after it stops. If Max drops from −25 dB to −45 dB, you have recorded a 20 dB swing in your room, even without certified units.
Tips for Better Readings
Keep the microphone still and uncovered — a finger over the port or a tight case can shift readings by several decibels.
Measure the change, not the number: compare before/after or place-to-place readings taken with the same device, weighting, and response.
Use Slow response for steady sources like fans and Fast for short events like door slams.
If you can borrow a calibrated meter, set both side by side and adjust the offset until this tool matches it — NIOSH found that calibrated references dramatically improve phone measurements.
Silence notifications before measuring; a single chime can take over the Max and Peak values.
Audio is analyzed live in your browser and never recorded, stored, or sent anywhere. The microphone stays off until you press Start and is released the moment you stop or leave the tab.
Was this tool helpful?
Want to tell us more?
0/500
Want us to follow up?
Thanks for your feedback!
Frequently Asked Questions
Why do the readings differ from a real sound level meter?
Browser and phone microphones are built for voice, not measurement: each model has its own sensitivity and frequency response, and none is calibrated to a reference pressure. NIOSH researchers who tested smartphone sound measurement apps found only a few came within ±2 dB of professional instruments, even under laboratory conditions.
This tool is honest about that: it reports levels relative to your device's full scale instead of pretending to show certified dB SPL. Differences and changes are meaningful; single absolute numbers are not.
Can this tool measure absolute dB SPL?
No — and no uncalibrated browser tool can. Absolute sound pressure level requires a calibrated measurement chain, which is what IEC 61672 class 1 and class 2 sound level meters provide.
What this tool measures well is relative level: how much louder one place, moment, or source is than another on the same device. For workplace, legal, or health assessments, use a calibrated class 1 or class 2 meter.
How do I use the calibration offset?
Place your device next to a trusted reference — ideally a calibrated sound level meter — play a steady sound such as constant fan noise, and adjust the offset in whole decibels until this tool matches the reference. The offset is remembered on your device and applied to all readings, including Min, Max, and Peak.
NIOSH follow-up research showed that anchoring smartphone measurements to calibrated equipment removes much of their variability. The offset makes readings comparable to your reference under similar conditions, but it still does not turn this tool into a certified instrument.
What do dB and dBFS mean here?
A decibel (dB) always expresses a ratio between two levels. Certified meters use dB SPL, referenced to a standard sound pressure in air. This tool uses dBFS — decibels relative to full scale — where 0 dB is the loudest signal your microphone path can deliver, and quieter sounds read as negative numbers.
With a calibration offset applied, the numbers shift toward your reference scale, but they remain relative measurements of what your particular microphone hears.
What is the difference between A and Z weighting?
A-weighting reshapes the measurement with the curve standardized in IEC 61672 to approximate how human hearing sensitivity varies with frequency: it discounts deep bass and the highest frequencies. It is the standard weighting for environmental and occupational noise, written dB(A).
Z-weighting (zero weighting) is flat — every frequency counts equally. Choose Z when you want the raw balance, for example when low-frequency rumble from traffic or machinery matters.
Should I choose Fast or Slow response?
Fast and Slow are the two time weightings IEC 61672 defines for sound level meters: exponential smoothing of the signal power with a 125 ms time constant (Fast) or a 1 second constant (Slow).
Use Fast to follow short events — door slams, barks, passing cars. Use Slow for a steadier reading of continuous sources such as ventilation, traffic hum, or machinery.
Is my audio recorded or uploaded?
No. The microphone signal is analyzed instantly in your browser to compute the level numbers, and the audio itself is never stored, recorded, or transmitted anywhere.
The microphone only activates after you press Start, and it is released the moment you stop, close the page, or switch away from the tab.
Why does the tool turn off noise suppression and other processing?
Browsers normally polish microphone audio for calls with echo cancellation, noise suppression, and automatic gain control. Automatic gain would quietly change the very levels this tool is trying to measure, so all three are requested off while measuring.
Your operating system or headset may still apply processing of its own — one more reason readings stay relative rather than absolute.
My Favorites
Drag to reorder
No favorites yet
Tap the ☆ on any tool page to bookmark it for quick access.