VID · Video tools

Remove Background Noise from Video

Shareable link

Settings are written to the URL as you change them. Nothing differs from the defaults yet.

Clean up a video recorded in a noisy room, next to a fan, or on a phone microphone with hiss. The steady noise under the speech is turned down while the voice stays, and the picture is not touched.

How the noise is found

The soundtrack is split into 28 frequency bands, and each band keeps a running estimate of its noise floor: it drops immediately when the band gets quieter and rises only slowly. Sound near that floor is turned down by up to the Reduction setting; sound well above it, such as speech, passes. Sensitivity sets how close to the floor sound must be to count as noise.

Electrical hum

Hum is a single frequency and its harmonics, so it is removed with narrow notches rather than the adaptive floor. Turn on Remove electrical hum, set 50 or 60 Hz, and raise Harmonics if a buzz remains.

Limits

Noise that changes from moment to moment, such as crowd chatter, music, or wind gusts, has no steady floor and survives. Very loud noise that masks the voice cannot be separated from it without damaging the speech. For picture noise, use the video denoiser instead.

Frequently Asked Questions

Steady ones: hiss, fans, air conditioning, traffic rumble, and room tone. It follows a noise floor that it measures as the video plays. Sudden noises, such as a door, a dog, or speech in the background, are not steady and pass through.

No. The noise floor is tracked per frequency band while the soundtrack plays. It needs at least one quiet moment, a pause or a breath, to learn where the floor sits.

When there is a low, constant buzz from mains power: bad cables, dimmers, or cheap lights. Set the frequency to 50 Hz for recordings from Europe, Africa, or most of Asia, and 60 Hz for the Americas.

Heavy reduction also removes quiet parts of speech that sit near the noise floor. Lower Reduction, or raise Preserve highs, until the voice sounds natural again.

Nothing. The video stream is copied packet for packet, so the picture keeps its exact quality and the export finishes in a fraction of the time a re-encode takes. Only the soundtrack is processed and re-encoded, and a note under the export button says which audio codec was written.

Explore Our Tools

Browse all tools