RTranslator is an offline real-time translation app for Android that enables live conversations between users speaking different languages.
The tool solves the problem of language barriers in face-to-face communication by combining speech recognition, translation, and text-to-speech entirely on the device. In Conversation mode, two or more users connect via the app; when one person speaks, their phone captures audio via microphone or Bluetooth headset, converts it to text, sends it to the other user's phone, translates it into their language, and plays it back as synthesized speech. WalkieTalkie mode works on a single phone, listening simultaneously in two languages, detecting which language the speaker uses, and translating and playing back the response. The app also includes a standard text translation mode. The tool uses Meta's NLLB model for translation and OpenAI's Whisper for speech recognition, both running locally on the device to ensure privacy and offline functionality.
Developers should choose this tool if they need real-time multilingual conversation support without cloud dependencies or privacy concerns. It suits projects requiring peer-to-peer communication with automatic language handling, particularly mobile applications where users may not share a common language. The Conversation mode is designed for extended dialogues and supports multiple simultaneous connections, while WalkieTalkie mode targets brief exchanges and works with a single device. The app can operate in the background, though some devices may throttle performance in this state. The tool's reliance on local AI models means translation quality does not degrade offline, but device computational resources will affect real-time performance.
Development activity shows consistent engagement with the codebase through regular commits addressing both features and maintenance. The project maintains active responsiveness to issues and pull requests, indicating ongoing attention to user-reported problems and community contributions. Documentation is kept current alongside code changes, reflecting a commitment to keeping the README aligned with actual functionality.
