OwOCR

OwOCR continuously monitors image sources such as the clipboard, screen captures, folders, and sockets, applying optical character recognition to extract text. While it is optimized for Japanese, it also supports many other languages and can output results to the clipboard, text files, or websockets. The tool integrates with Windows, macOS, and Linux, providing a tray icon for quick access and notifications, and can be controlled via keyboard shortcuts or a configuration interface.

Users can select among several OCR engines, including pre‑installed options like Bing, Apple Vision, and Live Text, and can add online services such as Google Lens for a two‑pass workflow that balances speed and accuracy. Features include furigana filtering for Japanese, capture of specific screen areas or windows, and support for Unix domain sockets on macOS/Linux.

Installation is available through native packages for Windows and macOS or via Python’s pip, with a configuration file stored in the user’s config directory. The application runs in the background, allowing users to pause, switch engines, or adjust settings from the tray icon or terminal.