Remove noise from audio
Drop in any audio file — voice memo, interview recording, narration — and get a cleaner version back.
Balanced (recommended)
Drop a video or audio file to clean up
or click to browse
How it works
Drop a video or audio file in. WavyVid decodes the audio track right in your browser.
A noise-suppression neural network (RNNoise) runs over the audio in 10ms frames, trained specifically to separate voice from background noise.
Listen to the before/after result, then download — video files get the cleaned audio muxed back in without re-encoding the picture.
When you'd use this
- You recorded a voice memo on your phone and it picked up more room noise than you noticed at the time.
- You have a field interview recording with wind, traffic, or crowd noise underneath the dialogue.
- You narrated a voiceover track and want it broadcast-clean before dropping it into a video edit.
Frequently asked questions
What audio formats are supported?
Most common formats — MP3, WAV, M4A, AAC, and more — since the underlying engine reads a wide range of audio containers and codecs.
How long does cleanup take?
Roughly real-time or faster on a modern laptop — a 10-minute recording usually finishes well under a couple of minutes.
Can I adjust how aggressive the cleanup is?
Yes — light, medium (recommended default), and aggressive strength options are available before you start.
Is my audio uploaded anywhere during this?
No — everything happens locally in your browser via WebAssembly. Nothing is sent to a server.
Does this work on music, or just voice?
It's tuned specifically to preserve speech, so it works best on voice recordings. Running it on music can dull higher frequencies the model mistakes for noise.