Background noise test
Hear your microphone. Check whether background noise gets in the way of your voice.
Stay quiet for 30 seconds, then speak for 10 seconds. Listen back and see whether background noise competes with your voice.
The test records audio for local playback only. Recordings are cleared when you start again or leave this page. Nothing is uploaded.
Set up your microphone and read dBFS
A repeatable fan comparison
- Put the microphone in your usual call position. Keep the same input volume and microphone for both tests.
- Label Test A “Fan on.” Stay quiet for 30 seconds, then read the sample sentence naturally for 10 seconds.
- Switch the fan off. Choose “Compare a change,” label Test B “Fan off” and repeat from the same seat.
- Compare the average background levels, then listen to both background and voice. Repeat if a passing car, keyboard or other one-off sound affected either sample.
Illustrative result: −38 dBFS with the fan on and −46 dBFS with it off means 8 dB less captured background level. These example numbers are not a prediction or a room sound-level measurement.
What does a negative number mean?
dBFS expresses the digital signal level relative to full scale. This tool calculates 10 × log10(mean sample power), referenced to amplitude 1. More negative values mean less captured signal: −50 dBFS is lower than −30 dBFS.
Turning down microphone gain also lowers the reading. A low number alone does not establish a quiet room or clear speech. Use the same setup, check the voice-to-noise estimate and listen back.
The average reflects all captured sound. The baseline describes the quieter part of your sample. The loudest moment shows the strongest 100-millisecond window.
The tool requests that automatic gain, echo cancellation and noise suppression be disabled. Your browser or hardware may still apply processing; review the notes in your results. A call app may process the same microphone differently.
Microphone troubleshooting
No permission prompt, or access was denied
Open this page directly over HTTPS and choose “Enable microphone.” Check the browser’s site permissions and your operating system’s microphone privacy settings. Allow access for this browser, then retry. A previously blocked permission may need changing in site settings before another prompt can appear.
The microphone is missing or the wrong input is selected
Allow microphone access first; browsers can restrict the available device list until permission is granted. Check that the input is connected and enabled in system sound settings. Stop the test before changing the microphone selector. If a saved selection is no longer available, choose “System default.” If it cannot open, close another app using the input and retry.
The meter stays empty, or the result warns about clipping
Speak briefly during the preview to check that the meter responds. Check hardware mute switches and system input volume if it does not. “Below measurement floor” means no measured energy, which can also indicate a muted or inactive input. If clipping is reported, reduce input gain or move farther from the microphone and repeat both tests with the new setup.
The test stops before it finishes
Keep this page visible throughout the quiet and speaking steps. Hiding it, disconnecting the input or suspending audio interrupts the test. Reconnect the microphone if needed and start again. Incomplete measurements are excluded from comparisons.
This test cannot convert dBFS into calibrated dB SPL, certify a room or assess hearing safety. It measures your microphone’s signal and provides a practical comparison. Audio stays on this device for local playback.
Browser references: microphone access and errors and available device permissions.
Privacy, exports and AI use
Audio is recorded in memory on this device for playback. Nothing is uploaded. Recordings and results disappear when you start again, refresh or leave this page. JSON downloads include measurements and processing notes, but exclude microphone names and device identifiers.
Share an exported JSON report with an AI to interpret your results. This tool does not yet offer an audio-upload API or MCP endpoint.
What it does
Check microphone background noise before a call, compare changes and export a report. Audio stays on your device.
How to use it
- Allow microphone access and check the preview. Run a 30-second quiet measurement followed by a 10-second speaking step.
- Change one thing and repeat with the same microphone position and gain. Requires HTTPS and AudioWorklet support.
- Use the troubleshooting guide for blocked permissions, silent inputs or clipping. Compare dBFS readings only within a controlled setup; they are not calibrated room-noise levels.
Example
Measure with a fan running, then turn it off and repeat with the same microphone position and input gain.
Methodology
Continuous 100 ms power windows measure relative digital audio level in dBFS, with a quiet sample and a speaking step.
Limitations
Not calibrated dB SPL or a hearing-safety assessment. Device processing, clipping and input changes can invalidate comparisons.
Automation
This tool runs in your browser. No upload API or MCP endpoint is available.
View capabilities and documentationReference updated
Join the conversation
Helpful tips, questions, and ideas for making this tool better.
Loading discussion…