LiveCaptions Translator is a real-time speech translation tool that integrates translation APIs with Windows LiveCaptions.
The tool solves the problem of translating live audio without requiring specialized hardware or opening multiple windows. It works by automatically invoking Windows' built-in LiveCaptions, which handles speech recognition, and then pipes the recognized text through a translation API of the user's choice. This approach leverages Windows LiveCaptions' resource efficiency and high recognition accuracy while adding translation capabilities through various backend services.
Developers should choose this tool if they need real-time speech translation on Windows systems without a Copilot+ PC and want to avoid the overhead of running separate applications. It suits scenarios like gaming, video watching, and live streaming where an overlay subtitle display is valuable. The tool supports both traditional translation services and LLM-based engines, with the README recommending LLM approaches for their ability to handle incomplete sentences and context. Available translation backends include Ollama, OpenAI-compatible APIs, OpenRouter, Google Translate, DeepL, Youdao, Baidu Translate, MTranServer, and LibreTranslate, offering both self-hosted and online options.
The project shows active development with regular feature additions and refinement of the integration between LiveCaptions and translation services. The codebase demonstrates attention to user experience through implementation of modern UI patterns, automatic theme switching, and thoughtful design choices like hiding LiveCaptions by default after first use. Documentation includes both English and Chinese language support, indicating consideration for an international user base.
