DeepLook
DESKTOP APP
DeepLook, in a native window
Live Studio, Transform, Audio Studio and the voice changer — the full app, on your desktop. Same account, same credits. Updates arrive automatically.
Version 1.0.11 · released 7 October 2026
Voice changer requirements
The real-time voice changer is a desktop-app feature on Windows and Linux. It runs on your own graphics card, not in the browser. Your graphics card is detected automatically; computers below the minimum can't use it.
For smooth use you need
- The DeepLook desktop app on Windows 10/11 (64-bit) or Linux.
- An NVIDIA graphics card with 6GB+ video memory. A laptop RTX 3060 is the minimum; an RTX 3070 / 4060 or better (8GB+) runs more comfortably.
- An up-to-date NVIDIA driver (version 570 or newer). Older drivers can't run the voice engine.
- 16GB of RAM and about 15GB of free disk space.
- Internet for the first start: the voice engine and voice models download once (about 7GB).
- On a laptop, keep it plugged in. On battery the graphics card slows down and your voice lags further behind.
- Headphones, and a headset or a microphone close to your mouth.
- On Windows, Smart App Control turned off (Windows Security → App & browser control). It blocks the voice engine from starting.
Known limits
- Your changed voice plays about half a second to a second after you speak.
- On speakers instead of headphones, echo cancellation can garble some words.
- In a noisy room, or far from the mic, more words come out wrong.
- The accent comes from the target voice: use a clear 10–25 second recording of one person talking.
- Now and then a single word may come out wrong.
- Not available in the web browser, on macOS, or on phones.
First launch
These builds aren't code-signed yet, so your OS will warn you the first time:
- Windows — SmartScreen says "Windows protected your PC". Click More info → Run anyway.
- Linux — mark the AppImage executable (
chmod +x) or install the.deb.
Prefer the browser? Open DeepLook on the web. · Trouble installing? Contact support.