Feature
Remove background noise from audio
Clean up voice memos, podcasts and interviews in seconds. The cleaner strips hiss, hum and room noise while keeping the voice intact, then levels the result to a consistent volume.
- Free with an account
- No card needed
- Up to 50 MB
- Returns a WAV
Cleaning runs on our servers, so you'll need a free account — no card required.
What it does
Smart denoising
Spectral subtraction learns the recording's constant background — hiss, hum, fan noise — and removes it without hollowing out the voice on top.
Auto-leveling
An EBU R128 loudness pass brings quiet recordings up to broadcast-standard volume, so a whispered memo lands at the same level as everything else.
Deleted after processing
Your upload is written to a temporary folder on the server, cleaned, and deleted as soon as the result has been sent back. It is never added to your library or used to train anything.
How to clean your audio
Upload your file
Select any audio or video file (MP3, WAV, M4A, MP4) up to 50 MB. The tool works best on speech — podcasts, interviews, voice memos — where there is a steady noise floor to subtract.
Adjust the strength (optional)
The default noise reduction of 85% suits most recordings. Higher settings strip more hiss but start to thin the voice and can add a metallic edge, so raise it only if noise survives.
Download the cleaned audio
Press process and download a WAV in seconds. WAV is uncompressed, so nothing is lost to a second round of lossy encoding before you edit it.
Questions
Do I need an account?
Yes — cleaning runs on our servers, so you need a free LrcSong account to use it. Signing up takes a moment and no card is required.
What happens to my audio?
Your file is uploaded, written to a temporary folder on the server for processing, and deleted as soon as the cleaned result has been sent back to you. It is never added to your library or used to train anything.
Which files can I clean?
Any common audio or video file — MP3, WAV, M4A, MP4 and similar — up to 50 MB. The audio track is extracted from video automatically.
How strong should the noise reduction be?
The 85% default suits most speech recordings. Push it higher only if hiss survives: aggressive settings start to thin the voice and can add a metallic edge.
What do I get back?
A cleaned WAV file. WAV is uncompressed, so nothing is lost to a second round of lossy encoding before you edit it.