Download
Download Kajo Voice
Local AI transcription for every platform. No account required for the free tier. Audio never leaves your device.
Coming soon
Native builds for macOS, Windows, and Linux are not available yet. Leave your email and we will send one message the day they are.
Join the waitlistSystem requirements
- OS: macOS 13 or later, Windows 10 (21H2) or later, or a modern Linux desktop (Ubuntu 22.04+, Fedora 39+, Arch, or any distro that runs AppImage).
- RAM: 16 GB minimum (Gemma 4 12B). Comfortable for transcription, library search, and cited chat.
- Disk: ≈2.2 GB of models before your first transcript (≈2.5 GB on macOS); ≈9 GB in total once the chat model finishes downloading in the background. Download once — everything runs offline afterwards.
- Network: optional after install — transcription, library, and cited chat work offline.
Frequently asked questions
Does Kajo Voice run on Linux?
Yes. Kajo Voice ships a native Linux build — .deb (Ubuntu/Debian), .rpm (Fedora/RHEL), and AppImage (any distro), for x86_64 and ARM64. Everything the Mac and Windows builds do runs on Linux.
Which Linux distributions are supported?
Ubuntu 22.04+, Debian 12+, Fedora 39+, RHEL/CentOS Stream 9+, Arch Linux (rolling), and any distro that can run AppImage. Both x86_64 and ARM64 builds are available.
Does Kajo Voice work offline on Linux?
Yes. Transcription, summarization, the searchable library, and cited chat all work offline on Linux — entirely on your hardware. No audio ever leaves your machine.
Does Kajo Voice support Wayland on Linux?
Yes. Kajo Voice supports both X11 and Wayland. Import and transcription do not require a special capture pipeline. System tray support uses AppIndicator/StatusNotifierItem.
Waitlist
Be first to know
Kajo Voice is not available yet. Leave your email and we will send one message the day it is.