Canonical has introduced Myna, an AI-powered speech-to-text tool for Ubuntu desktop available as a technology preview.
Built as a collection of snaps including an orchestrator, GUI setup, and local AI models, Myna processes audio entirely in memory to dictate into any text field on GNOME and Wayland environments across selected Ubuntu releases.
The tool, called Myna, allows users to dictate text into any typing field on their computer. The feature is based on local artificial intelligence, which means the audio is processed exclusively in the device's memory and deleted immediately, without being stored or sent off the computer.
How it works and what it offers
When the user loses focus on the field they are writing in, the recorded speech is lost instead of being pasted elsewhere. The trial version offers a choice of three different open voice-to-text models: Nvidia's Parakeet, OpenAI's Whisper, and Alibaba's Funasr.
The tool consists of the main program Myna, a configuration tool with a visual interface, and the artificial intelligence model itself.
Who can try it?
The feature is available for testing in Ubuntu 24.04 LTS, 26.04 LTS, and 26.10, using the default GNOME and Wayland desktop environments. Canonical plans to add support for other desktop environments in future releases.
The company says the feature is under active development and features may change or experience issues in the coming months.
Top comments (0)