Skip to content

Noise Remover

Clean up recordings by removing background noise, hums, and unwanted sounds. Perfect for podcasts, interviews, and voice-overs.

Supports MP3, WAV, OGG, FLAC and more

How to use Noise Remover.

4 simple steps to get started.

1
Step 1

Upload audio

Select the noisy audio file.

2
Step 2

Detect noise

The tool automatically detects background noise.

3
Step 3

Adjust level

Fine-tune the noise reduction intensity.

4
Step 4

Export clean audio

Download the cleaned-up audio file.

Why use Noise Remover?

Everything you need to know about noise remover.

Noise Remover — free & online

Clean up recordings by removing background noise, hums, and unwanted sounds. Perfect for podcasts, interviews, and voice-overs. No downloads, no sign-ups, and no accounts. Editing and preview run locally in your browser using the Web Audio API; exporting sends your file over an encrypted connection to our processing server, which deletes it as soon as your download finishes.

How does Noise Remover work?

Simply upload your audio file — MP3, WAV, OGG, FLAC, or M4A — and use our intuitive editor to make your changes. Upload audio, Detect noise, Adjust level, Export clean audio. Once you're satisfied, export the result in your preferred format instantly.

Who is Noise Remover for?

Whether you're a podcaster, musician, content creator, or just someone who needs a quick audio edit, Noise Remover is built for you. It works on any device — desktop, tablet, or mobile — across all modern browsers including Chrome, Firefox, Safari, and Edge.

What can and cannot be removed

This works by spectral noise reduction. It builds a profile of the steady background sound in the recording, then subtracts that profile from every moment of the audio. The consequence is important: it is very effective against noise that is constant and unchanging, and close to useless against noise that is not.

  • Removes well — tape hiss, mains hum, computer fans, air conditioning, distant traffic, the general room tone of a quiet office.
  • Removes poorly or not at all — a door slamming, a dog barking, someone else talking, a chair scraping, wind gusts. These are events rather than a constant floor, so there is no stable profile to subtract.

For one-off noises, cutting them out with the MP3 Cutter is the realistic fix. Turn on Remove Selection, highlight the cough or the bang, and the audio either side is joined back together.

How much reduction to use

The reduction control runs from 0 to 100%. More is not better. Noise reduction works by removing spectral content, and voices share spectrum with the noise, so aggressive settings take part of the voice with the hiss. That is what produces the underwater, robotic quality familiar from over-processed recordings.

Start around 40 to 60%. That is usually enough to push background noise below the point where you notice it while leaving the voice intact. Push higher only if the noise is genuinely loud, and listen to consonants — s and t sounds degrade first and are the earliest warning that you have gone too far.

A recording with audible noise and an intact voice is better than a silent-background recording with a damaged voice. Listeners tolerate background hiss far more readily than processing artefacts.

Where it fits in a clean-up

Order matters. Reduce noise before compressing, not after: the Audio Compressor raises quiet passages, and any noise still present gets raised along with them, so compressing first makes the noise harder to remove. A sensible sequence is trim the dead air, reduce noise, compress to even out the levels, then adjust overall volume last. Processing happens on the server at export and the uploaded file is deleted immediately afterwards.

Frequently asked questions.

Everything you need to know about Noise Remover.

What types of noise can it remove?

Steady, continuous noise: air conditioning, fan hum, mains buzz, tape hiss, distant traffic. It profiles the constant part of the signal and subtracts it, which means one-off sounds — a door slam, a cough, a chair creak — stay exactly where they are.

Will it affect voice quality?

At moderate settings the effect on speech is small. Push the reduction high and voices start to sound thin, watery or metallic, because the process also removes the parts of the voice that share frequencies with the noise.

Can I adjust the noise reduction level?

Yes, from 0 to 100%, with 50% as the default. That figure is how much of the detected noise is subtracted, not a quality setting — higher is not better.

Does it work with music?

It is built for voice. On music it can tame hiss on an old recording, but cymbals, reverb tails and quiet detail occupy the same range as the noise, so they get thinned along with it.

How do I remove background noise from a voice recording?

Start at 40–50%, export, and listen to the gaps between words — that is where artefacts appear first. If noise is still audible, raise the amount in steps of ten rather than jumping straight to 100.

How much noise reduction is too much?

Once the voice takes on a hollow, underwater quality you have gone past the useful point. Leaving a little steady hiss in place is nearly always better than stripping the voice, because listeners tolerate hiss far more readily than warbling artefacts.

Can it remove echo or reverb from a room?

No. Echo is your own voice arriving late rather than steady background noise, so there is no constant profile to subtract. Reverb has to be dealt with when recording — soft furnishings, a closer microphone, or a smaller room.

Can it remove music from behind a voice?

No, because music changes constantly and never settles into a steady noise profile. To take the vocal out of a song, use the Karaoke Maker, which works on the stereo image instead.