Pot

Pot is an open-source desktop app that provides text translation and optical character recognition (OCR) across multiple platforms. It allows users to translate text via selection, manual input, or screen capture, making it suitable for users who frequently read foreign language content on their computers. The application functions as a utility layer that sits on top of the operating system to provide quick access to linguistic tools without switching windows.

The application is available for Windows, macOS, and Linux, including support for Wayland environments such as KDE, Gnome, and Hyprland. It can be installed via package managers like Winget, Homebrew, and AUR, or through manual installation packages. Users can configure specific keyboard shortcuts to trigger different translation modes or enable a clipboard listener that automatically translates any text copied to the system clipboard.

Key features

• Parallel translation using multiple API interfaces
• Text recognition via OCR with offline system support
• Text-to-speech synthesis for translated content
• Exporting translated terms to vocabulary books like Anki
• Screen capture for direct OCR and translation
• Clipboard monitoring for automatic translation upon copying
• Plugin system for adding external translation services
• HTTP API for integration with other software

Built with Tauri, the application integrates with various AI and translation services including OpenAI, Gemini Pro, and Ollama for offline use. It supports a wide array of OCR engines, utilizing Windows.Media.OCR, Apple Vision Framework, and Tesseract depending on the operating system. For advanced users, the software provides a full HTTP interface on a local port, allowing it to be called by external scripts or third party tools. This architecture enables integration with productivity extensions like PopClip on macOS or SnipDo on Windows, which can trigger Pot translations directly from a selection menu.

Pot is a versatile, cross-platform utility for users who require integrated translation and OCR capabilities within their desktop workflow.