Remove any unwanted
sound from your videos
Just describe the sound in plain words - remove it, remove everything else, or adjust its volume.
See it in action
Check out the results below ↓Before & After
Hear the difference for yourself
Background Music Removed
Copyrighted music playing in the background? Remove it and keep your footage safe from demonetization.
Speech Removed
Someone talking over your footage? Remove the speech and rescue the video you actually wanted.
Ambient Noise Removed
Wind, traffic, crowd chatter - remove environmental noise from moments you can never re-film.
Built for real problems
Not just a tech demo - solving actual pain points creators face every day
Travel creators
Filmed at a cafe with copyrighted music playing? A street performer in the background? Remove the music and keep your footage - no more demonetization strikes.
Unwanted commentary
Great footage ruined by someone talking over it? Inappropriate comments caught on camera? Remove the speech and rescue the video you actually wanted.
Irreplaceable moments
Wedding footage with an awkward remark? A baby's first steps with TV blaring? Remove the noise from moments you can never re-film.
Target any sound
Remove it, remove everything else, or adjust the volume of each - your choice
Speech
Remove voices, keep only the dialogue, or adjust the volume balance between speech and everything else
Music
Remove background music, isolate just the music track, or dial the volume up and down independently
Ambient Noise
Remove wind, crowd noise, and environmental sounds - or adjust how much ambience stays in your mix
Animal Sounds
Remove barking, chirping, and other animal sounds - or turn them down without removing them entirely
Urban Noise
Remove traffic, construction, and sirens - or control exactly how much bleeds through
Custom
Describe any sound in plain text - remove it, remove everything else, or fine-tune the volume of each layer
Drag & Drop
Simply drag your video or audio file in - no complex setup required
Real-time Progress
Track processing status with a live progress bar
Job History
Track all your processed jobs and access output files locally
Building an app? Use the SAM Audio API to integrate audio source separation directly into your product.
Your videos never leave your computer
We only extract and process the audio track - your video files stay on your machine the entire time. We never store your audio. Once processing is done, it's gone from our servers.
Frequently Asked Questions
Everything you need to know about SoundScrub
What types of sounds can SoundScrub handle?
SoundScrub can target a wide variety of sounds in your videos, including speech/dialogue, background music, and ambient noise. You can also use custom prompts to target specific sounds like "dog barking", "keyboard clicking", "traffic noise", "crowd chatter", or any other sound you can describe in natural language. For each sound, you can remove it completely, remove everything else, or use adjust mode to control the volume of the targeted sound and everything else independently.
How much does it cost to process a file?
SoundScrub uses a simple credit-based pricing: $0.25 per 30 seconds of footage. For example, a 5-minute file costs $2.50, and a 30-minute file costs $15.00. There are no subscriptions or hidden fees - you only pay for what you process. Credits never expire and can be used whenever you need them.
What video formats does SoundScrub support?
SoundScrub supports all common video formats including MP4, MOV, AVI, MKV, WebM, and more. The app automatically extracts the audio from your video file, processes it, and returns a new video with the same quality and format as the original - only the unwanted audio is removed.
Can I use SoundScrub with audio-only files?
Yes. SoundScrub also accepts standalone audio files including MP3, WAV, FLAC, M4A, OGG, and OPUS. The processed result is saved in the same format as your input file. Pricing is identical to video - $0.25 per 30 seconds of audio.
Is my video data secure and private?
Yes. Your videos never leave your computer. Only the extracted audio is sent to our servers for processing, and it's deleted immediately after processing is complete. We don't store, analyze, or use your content for any purpose other than performing the requested audio removal. All communication between the desktop app and our processing servers is encrypted. We never see your payment details - they're handled securely by Stripe, our payment processor.
What platforms is SoundScrub available on?
SoundScrub is available for macOS (Apple Silicon), Windows (x64 and arm64), and Linux (x64 and arm64). The desktop app provides a native experience while leveraging cloud-powered AI for processing.
How long does it take to process a file?
Processing time varies based on file length and server load, but typically takes a few minutes for most files. The SoundScrub desktop app provides real-time progress updates, and you can continue using your computer while processing happens in the background.
Can I remove multiple sounds from the same file?
Each processing job removes one type of sound. To remove multiple different sounds from the same file, you can process the file multiple times - first removing one sound, then processing the result to remove another sound. Each processing job is charged separately based on file length.
What happens if I'm not satisfied with the results?
We offer a 7-day refund policy for technical issues or processing failures. If the audio separation doesn't work as expected due to a technical problem on our end, contact us at [email protected] within 7 days for a refund. Refunds do not cover cases where the AI separation quality doesn't meet expectations - complex audio mixes or unusual sounds may not separate perfectly, and that is an inherent limitation of the technology.
What AI model does SoundScrub use?
SoundScrub is powered by Meta's SAM Audio (Segment Anything Model for Audio), a state-of-the-art AI model for audio source separation. SAM Audio can isolate and remove specific sounds from complex audio mixes using natural language descriptions. SoundScrub makes this model accessible as a simple desktop app - no Python, no GPU setup, no command line required.
