We saw all your comments about Linux... and we're beyond stoked to finally bring it to you. So what's next for everyone to ask for now? Superwhisper for smart fridge?
Kidding. Kidding.
Getting voice input working on Linux has always meant stitching together recording tools, shell shortcuts, and clipboard hacks that break the moment you change compositors.
When you work in a terminal, editor, or browser, typing a full explanation takes time. You know the exact context of a bug, a PR, or an agent prompt, but shorten it to save time. Then you have to explain it again anyway. Talking gets that context down while it is still fresh.
We built Superwhisper for Linux from scratch to make system-wide voice input fast, native, and reliable.
Native voice input, zero cloud required
- Local or cloud processing: Run models locally on your own CPU or GPU so your audio never leaves your hardware, or switch to cloud processing when you want maximum speed without running a model on your machine
- Built for keyboard-driven workflows: Configure everything through a clean graphical interface, a dedicated terminal UI, or directly via CLI commands
- Seamless text insertion: Text lands in your focused app without switching windows
Set a shortcut and start talking. Superwhisper types the cleaned-up text wherever your cursor is.
Supported distributions
Superwhisper for Linux launches on all the major distributions: Omarchy, Arch, CachyOS, Fedora, Ubuntu, Debian, and openSUSE, on Wayland. That's just where we're starting. We want Superwhisper to work on whatever distro you actually use, so if yours isn't on the list, tell us. Send us a message on X at @superwhisper if you want us to test something specific on your setup, or if you'd be up for helping us get it running there.
For Linux updates and help with your setup, join our Discord.