Discord Mic Test: Test Your Microphone Before You Talk

Test Your Microphone in a Browser Before Discord

When Discord's bar stays flat, you don't yet know whether the hardware or the app is at fault. An online microphone test answers that: it runs in your browser, needs no download, and works as a free mic test you can repeat freely. Most tools process audio locally, so nothing leaves your computer unless you save a recording yourself. Try the free earbuds and airpods test for a quick, no-install way to check this yourself.

Allow Microphone Permission and Watch the Waveform

Click the start button, then click Allow when the browser asks for access. Speak normally: a moving waveform means the microphone works at the hardware level. Many also list the device name and sample rate, handy when two microphones share similar names.

Hear Yourself Through the Device Selector

Use the device selector to switch between a gaming headset, a webcam microphone or AirPods and check each one in turn. Then listen to the mic test playback or a short recording: hiss and a muffled tone show up on playback far more clearly than on a meter. If that sounds fine but Discord still fails, the problem sits inside the app.

How I Run the Test Before a Tabletop Night on Discord

I'm running a five-player tabletop campaign on Discord with a new USB condenser mic, and last session two players said my dice rolls were drowned out by keyboard clatter. The decision I need to make tonight: move the mic closer or turn the gain up. The answer is to move it, and here is how the numbers showed that.

Discord's own check only confirmed the mic was detected, so I opened a level meter, set the mic 30 cm from my mouth, read a paragraph of session notes, then typed a sentence on the mechanical keyboard. The readings:

At 30 cm my speech averaged well below the −18 dBFS alignment level that EBU R68 uses as its reference, and only 13.9 dB separated speech from the keys. Halving the distance adds 20 × log10(30 ÷ 15) = 6.02 dB to direct speech, which I confirmed with a second reading at 15 cm. The keyboard sits where it was, so the gap to clatter grows to 19.9 dB. Turning gain up 6 dB instead lifts everything together: speech looks identical, but the clatter climbs to −35.3 dBFS and the gap stays 13.9 dB.

The number I'd watch is the keyboard peak, not the speech average. With Discord's input sensitivity set manually to −35 dB, the 15 cm setup keeps every keystroke below the gate while speech clears it by more than 13 dB. The gain route puts the clatter at −35.3 dBFS, a fraction of a decibel under that line, so a heavier keystroke opens the gate anyway.

One caveat: the 6 dB rule only holds for a mic that isn't already very close. Under about 5 cm, the proximity effect boosts bass and plosives, so a microphone check at that distance overstates how clear you'll sound. My next step is a 15 cm boom arm and one more reading after the first dice roll.

Reading Results from an Online Mic Test

Discord's own test bar, like any pass/fail light, only tells you the microphone is detected, so a meter is what you use to set its input sensitivity slider. A meter that reports peak, average and floor levels in dBFS (decibels relative to full scale, where 0 is the ceiling) tells you how you'll come across. As a rule of thumb, speech averaging roughly −24 to −12 dBFS, with peaks a few decibels short of 0, sounds clear without distortion. Test your mic with the same headset and room you'll use on calls, or the reading means little.

Noise Floor and Headroom in One Example

Here's a headset reading to compare against. On the first pass, with the gain slider maxed and a desk fan running, the meter showed:

• Peak: 0 dBFS, so every loud word clipped
• Average speech: −10 dBFS
• Room floor: −46 dBFS, a 36 dB gap below speech

That's two problems: zero safety margin above the loudest words, and a room that is too loud. Lowering the gain by 9 dB moved all three readings down together, to −9, −19 and −55 dBFS. The gap stayed at 36 dB, because gain raises the noise floor and your speech equally. Switching the fan off then dropped the floor to −63 dBFS. Final result: 9 dB of headroom and a 44 dB gap between speech and room sound, so a Discord mic test afterwards plays back clean. If you run the test again after each single change, you'll know which change actually helped.