Why dBFS and not real decibels (dB SPL)?

A browser mic reports a digital signal level, not calibrated acoustic pressure. dBFS measures loudness relative to the maximum representable level. Real-world dB SPL needs a calibrated reference mic, which a generic web tool can't provide — so this meter is relative: great for checking the mic works and isn't clipping, not a calibrated SPL meter.

Why are device names blank until I start?

Browsers hide device labels until microphone permission is granted at least once, to prevent fingerprinting. After your first Start, the dropdown fills in with your actual device names.

Does the microphone test work differently in Chrome, Firefox, and Safari?

Broadly the same, but with small differences. Chrome and Edge may apply audio processing (auto-gain, noise suppression) even when the page requests them off — the Processing readout shows what the browser reports. Firefox generally honors the raw-signal request more consistently. Safari requires iOS 14.5+ or macOS 11+ for Web Audio mic access. The level meter and waveform work in all modern browsers; spectrum accuracy and exact levels may vary slightly.

🎤

Microphone Test

Run a quick microphone test online to check whether your mic works — no software to install, just click Start test in your browser. This free online mic test is the fastest way to do a mic check: allow microphone access, then speak or tap the mic and you’ll see a live input-level meter (RMS + peak in dBFS), a clipping warning, and a real-time waveform and spectrum. Pick a specific input device and read back its actual sample rate and channel count.

🔒 Your audio never leaves your device. Everything runs locally in your browser via the Web Audio API — nothing is recorded, uploaded, or stored.

Microphone

Input device

Device names appear after you grant permission once. Changing device restarts the test.

Idle — press Start test.

Verdict

Not started

Start the test, then make some noise. A working mic shows movement on the meter within a second or two.

Detected properties

Sample rate

—

Channels

—

Peak level

—

Processing

—

Input level

-60-48-36-24-120 dBFS

RMS —

Peak —

Clip —

Aim for peaks around -12 to -6 dBFS when speaking normally. If the Clip indicator lights, lower your input gain or move back from the mic.

Live waveform — your microphone signal in the time domain

Live spectrum — frequency content of your microphone signal (log scale)

How the Microphone Test Works

This tool uses your browser’s Web Audio API to read the raw signal from your microphone in real time. When you click Start test, the browser asks for permission to use the mic; once you allow it, the audio is routed into an AnalyserNode that measures the signal level and frequency content many times per second. Nothing is sent anywhere — the analysis happens entirely on your own device, and the audio is discarded as fast as it arrives. This approach works with any input your OS exposes to the browser: built-in laptop mics, USB microphones, XLR mics via audio interfaces, wireless headsets, and webcam microphones — making it equally effective as a webcam mic test or a quick way to test my microphone on any device.

The level meter shows two numbers in dBFS (decibels relative to full scale, where 0 dBFS is the loudest a digital signal can be without clipping): RMS is the average / perceived loudness, and Peak is the loudest instantaneous sample. A healthy speaking level peaks somewhere around -12 to -6 dBFS. If peaks reach 0 dBFS the signal is clipping — distorting — and you should reduce your input gain. If you want a deeper profile beyond just checking levels (including a noise floor and SNR measurement), try the mic sensitivity analyzer for a structured 3-step test.

What a mic test can and can’t tell you

Can confirm: the browser sees your mic, the device produces a signal when you make noise, the approximate input level, whether you’re clipping, and the reported sample rate / channel count.
Can’t confirm: absolute sound-pressure level in real-world dB SPL (that needs a calibrated reference — this is a relative dBFS meter), nor the subjective "quality" of your mic. Browser-applied processing (auto-gain, noise suppression) can also color what you see.

If your mic isn’t working

No permission prompt / blocked: check the padlock icon in your address bar and allow microphone access for this site, then press Start again.
Meter stays flat: pick a different device in the Input-device dropdown — your OS default may be the wrong mic (e.g. a disconnected headset).
Nothing at all: check your operating-system privacy settings (microphone access for the browser) and that the mic isn’t muted in hardware or the OS mixer.
Only works in one app: close other apps that may be holding exclusive access to the microphone.
Consistently quiet level: your input gain may be set too low. Raise the microphone level in your OS sound settings, then check the RMS and Peak readings again. For a structured way to find the right gain, use the mic gain level meter with its peak-hold display.

Frequently Asked Questions

Is my audio recorded or uploaded?

No. The test runs entirely in your browser using the Web Audio API. Your microphone signal is analyzed locally to drive the meter and graphs, and is never recorded, saved, or transmitted to any server. When you stop the test (or close the tab) the microphone is released immediately.

Why does it ask for microphone permission?

A browser can’t read a microphone without your explicit consent — that’s a privacy protection built into every modern browser. The permission prompt appears when you click Start test. You can revoke it at any time from the padlock/site-settings menu in your address bar.

What do RMS and Peak mean?

Both are measured in dBFS (decibels relative to digital full scale). Peak is the single loudest sample in the current frame — it tells you how close you are to clipping (0 dBFS). RMS is a running average that tracks perceived loudness more closely. For voice, aim for peaks around -12 to -6 dBFS with RMS noticeably lower.

What is clipping and why does it matter?

Clipping happens when the signal exceeds 0 dBFS and gets "flattened" at the top — it sounds harsh and distorted, and the lost detail can’t be recovered. If the Clip indicator lights up, reduce your microphone gain in your OS sound settings, move slightly farther from the mic, or speak more softly.

Why is the level meter in dBFS and not real decibels (dB SPL)?

A microphone in a browser reports a digital signal level, not a calibrated acoustic pressure. dBFS measures how loud that digital signal is relative to the maximum the system can represent. Converting to real-world dB SPL requires a calibrated reference microphone, which a generic web tool can’t provide. So this meter is relative — perfect for checking that your mic works and isn’t clipping, but not a substitute for a calibrated SPL meter.

Why are the device names blank until I start?

Browsers hide device labels until you’ve granted microphone permission at least once, to prevent fingerprinting. After your first Start, the dropdown fills in with your actual device names so you can switch between inputs.

Does it disable noise suppression and auto-gain?

It requests the raw signal (autoGainControl, noiseSuppression and echoCancellation all off) so the meter reflects what your mic actually captures. Some browsers/operating systems may still apply processing they don’t let the page disable; the Processing readout shows what the browser reports for the active track.

Can I hear myself (monitor)?

Yes — once the test is running, click Monitor (hear yourself) to route your microphone to your output so you can hear it live. It’s off by default. Use headphones: on speakers the mic picks up its own output and you’ll get a feedback squeal. Monitoring uses the same in-browser audio path — nothing is recorded or sent anywhere.

Can I test a headset, USB, or Bluetooth mic?

Yes — any input the browser can see appears in the Input-device dropdown. Select it and the test restarts on that device. Note that Bluetooth headsets often switch to a low-quality "hands-free" mode (8–16 kHz) when their mic is in use; the Sample rate readout will reflect that.

Why is the live spectrum flat or empty when I speak?

A flat spectrum almost always means the microphone isn’t actually sending signal to the browser — even if the level meter shows a little movement from OS processing. Common causes: the wrong device is selected in the Input-device dropdown (switch to a different one), your OS is routing the mic to a different app, or browser noise suppression is zeroing everything below a threshold. Try switching devices, disabling mic enhancements in OS sound settings, and re-starting the test.

Does this work for testing a built-in webcam microphone?

Yes. A webcam’s built-in microphone appears as a regular audio input device in your OS and in the browser’s input-device list. Select it from the dropdown (names usually include the camera model or "USB audio") and run the test normally. Webcam mics often have a high noise floor and narrow frequency range, so expect a lower SNR than a dedicated microphone — the level meter and spectrum will show this clearly.

Does the test work differently in Chrome vs Firefox vs Safari?

Broadly the same, but with small differences. Chrome and Edge tend to apply more aggressive default audio processing (auto-gain, noise suppression) even when the page requests them off — the Processing readout will say "browser-managed" if the page can’t override it. Firefox generally honors the constraint request more consistently. Safari requires iOS 14.5+ or macOS 11+ for Web Audio mic access, and may limit sample rates. The core level-meter and waveform work in all modern browsers; the spectrum and exact level accuracy may vary slightly.

Related Tools

← All Microphone Diagnostics Tools