Handy: A free, open source, and extensible speech-to-text application that works completely offline.
What problem it solves
Handy addresses the need for a speech-to-text tool that is private, open to extension, and simple to operate. It lets users dictate into any text field without sending audio to the cloud.
Key features
Offline speech transcription — Audio is processed on the user's computer, keeping the transcription workflow local and private.
Source: README
Direct text insertion — A configurable shortcut starts and stops recording, then pastes the resulting transcription into the active application.
Source: README
Choice of transcription models — Users can select from Whisper models or Parakeet V3, with GPU acceleration available for Whisper when the system supports it.
Source: README
Cross-platform desktop support — The application supports Windows, macOS, and Linux, with platform-specific installation and text-input guidance.
Source: README, BUILD.md
Extensible workflows — Advanced workflows include command-line control, a Raycast integration, and a transcription hook that can rewrite text before it is pasted.
Source: README
Who it may suit
Inference: may suit users who want private, keyboard-driven dictation on Windows, macOS, or Linux, as well as developers interested in extending an offline speech-to-text application.
Topics to discuss
Which offline transcription workflows and model-selection options are most important for your daily use?
How should the transcription hook and other extension points be documented for users and contributors?
What accessibility, Linux display-server, or text-input issues should the community prioritize next?
Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.
Read the original README https://github.com/cjpais/Handy#readme
ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.
Comments
…