You filmed on a street, in a cafe or in the wind and the speech got buried. Separation gives you two files: the dialogue alone, clean, and the environment alone, with the traffic, the crowd or the music. Use whichever you need.
- Price
- 1 credit per started minute
- Length
- Up to 20 minutes per file
- Privacy
- Runs on Cadabra's own server, the audio goes to no third party
- Input
- An audio or video file
- Billing
- The exact price shows before you confirm, and a failed run returns its credits
Three things to do with it
Cleanup: take the clean dialogue file and use it instead of the original. Dubbing: keep the environment file and lay a new voice or another language over it, so the scene still sounds like the place it was filmed. Transcription: transcribe the clean speech so the captions come out more accurate.
Why keep the environment
When you replace a voice and leave silence under it, it sounds fake at once. The original environment, the wind, the crowd, the room, is what makes the new voice sound as if it was spoken there. That is why separation returns both files and not only the speech.
What it costs
A 4-minute video costs 4 credits. The system reads the file's real length, and the price shows before you run it.
Common questions
Does it remove music under the speech too?
Yes. The music goes to the environment file and the speech stays alone.
Can I upload a video file?
Yes. Give a link to the audio or video file itself and the system takes the sound from it.
What happens if a generation fails?
A generation that fails for a technical reason returns its credits to your balance automatically. It is code that runs on every job, not a request you have to send.
Can I do this from inside Claude or ChatGPT?
Yes. Connect Cadabra once and ask in plain words in the chat. The chat shows the price, waits for your yes, and the result comes back to the same conversation and is saved to your gallery.
More on Cadabra
- The full price list for every tool and engine
- Automatic Hebrew captions
- Voice changer
- AI Video Generator, from Text or a Photo
- AI Image Generator, Every Top Engine in One Place
- AI Voiceover and Text to Speech
- Every AI Engine in One Place
- Credits, Not a Monthly Subscription
- AI That Works in Hebrew
- Cadabra AI, One Studio for Every AI Engine
- AI Animation Generator, from Character to Film
- AI Song and Music Generator
- A Video for a Small Business, with AI
- What It Costs to Make a Video with AI
- Turn a Photo into a Video, with AI
- Hebrew Captions for Your Video, Automatic and Free
- AI Video Upscaler to Real 4K
- AI Image Upscaler
- Remove a Video Background, No Green Screen
- Translate and Dub a Video in Your Own Voice
- AI Lip Sync
- AI Voice Cloning from 30 Seconds of Speech
- AI Voice Changer That Keeps the Performance
- Extend an Image to Any Shape, Without Cropping
- AI Thumbnail Maker for YouTube and Reels
- Turn Wide Video into Vertical Reels
- AI Sound Effect Generator
- AI Talking Avatar from One Photo
- AI UGC Videos from One Product Photo
- AI Product Video from One Photo
- AI Product Photography
- AI Video Ad Generator, from One Sentence
- AI Movie Generator, from One Idea
- AI Storyboard Generator, from Script to Video
- Consistent AI Characters, the Same Face in Every Shot
- AI Reels and Shorts Maker
- Make AI Videos Inside Claude
- Make Videos Inside ChatGPT
- AI Headshot Generator from One Selfie
- AI Action Figure from Your Photo
- Put Yourself in Any Video Clip
- Up to 10 Versions of One Ad
- New Camera Angles from One Image
- AI 3D World Generator
- AI Music Video, Cut to the Beat
- From Script to an Edited Video
- From a Product Link to a Video Ad
- AI Micro Drama, Episode by Episode
- Turn Photos into an Edited Video