Mic Test: Check Your Microphone Online

How to test your microphone

1. Choose a microphone if you have more than one, then press Start. Your browser asks for permission the first time; allow it.
2. Speak at your normal distance. The bar moves, the waveform draws your voice, and a green message confirms that sound is reaching the browser.
3. Press Record to capture a 5-second clip (or 10 or 30 seconds). It plays back automatically when it finishes, so you hear what the browser captures with the current filter settings. Download it compressed, as the browser recorded it, or as an uncompressed WAV file.
4. Press Stop when you are done. The page stops every audio track, so the browser’s recording indicator switches off.

Device names only appear after you grant access, because browsers hide them from pages that have no permission. If the level stays flat, try the other entries in the list: laptops often expose a built-in array, a headset and a “communications” device separately.

Reading the meter

Levels are shown in dBFS, decibels relative to full scale. 0 dBFS is the largest value digital audio can hold, so every reading is zero or negative. Peak is the loudest single sample in the last 2,048 samples (about 43 ms at 48 kHz), and Level (RMS) is the root-mean-square level over the same window. For a steady sine wave the RMS reading sits 3 dB below the peak; speech is spikier, so the gap is wider. Peak hold keeps the loudest peak since you pressed Start, the number to check before recording.

The fill colour follows the same bands: amber while the peak is below −25 dBFS, green from −25 to −6 dBFS, and red above.

Compressed or WAV download

The playback and the compressed download come from the browser’s recorder, usually WebM with the lossy Opus codec (M4A with AAC in browsers that cannot record WebM). The WAV file is 16-bit PCM of the exact samples the meter measured, at the rate in the settings table: use it to judge hiss or hum, or to edit the clip. It takes about 94 KB per second at 48 kHz mono.

Echo cancellation, noise suppression and auto gain

By default, browsers run microphone audio through the same kind of processing used for video calls. The three stages are the echoCancellation, noiseSuppression and autoGainControl constraints of the W3C Media Capture and Streams specification. The checkboxes above switch them on or off where the browser supports it, and the settings table shows what was actually applied. Automatic gain gradually raises quiet input and lowers loud input, so turn it off when you want to judge the real volume set in your operating system. Turn all three off reopens the microphone once with the rawest signal the browser allows; music apps do the same because the filters are tuned for speech.

When the microphone will not start

• Permission denied: click the icon at the left of the address bar, allow the microphone for this site and press Start again.
• Blocked by the system: on Windows open Settings, Privacy & security, Microphone; on macOS open System Settings, Privacy & Security, Microphone, and allow your browser.
• In use: another app may hold the device exclusively. Close call apps and other tabs that use the microphone.
• Not secure: browsers only allow microphones on https pages and on localhost.