sokuji

Real-time two-way speech translation for bilingual meetings — auto-detects the spoken language and translates both directions, cloud or fully offline on-device.

What is sokuji?

Real-time two-way speech translation for bilingual meetings — auto-detects the spoken language and translates both directions, cloud or fully offline on-device. Desktop (Windows · macOS · Linux) + browser extension (Chrome · Edge) for Zoom, Meet, Teams & any app.

Overview of sokuji

What it is

Sokuji is a real-time two-way speech translation application for bilingual meetings. It auto-detects the spoken language and translates in both directions. It can run via cloud providers or fully offline on the user's device using local inference powered by WASM and WebGPU. Sokuji is available as a desktop app (Windows, macOS, Linux) and a browser extension (Chrome, Edge). It is built by Kizuna AI Lab.

Key features

  • Local Inference (Edge AI): Runs ASR, translation, and TTS entirely on-device with no API key, no internet, and no GPU required. Includes 44 ASR models, 75 translation models, and 137 TTS models.
  • Cloud Providers: Integrates with 8 providers including OpenAI, Google Gemini, Palabra.ai, Kizuna AI, Doubao AST 2.0, Soniox, and OpenAI-compatible APIs.
  • Two-Way Translation: Auto-detects the language being spoken and translates into the other language in both directions simultaneously. Supports 60+ languages and 3,600+ language pairs via Soniox.
  • Audio Features: Virtual microphone to route translated audio to any app, real-time passthrough, AI noise suppression, and echo cancellation.
  • Interface: Fully localized UI in 30 languages, with Simple Mode for non-technical users and Advanced Mode with waveform display.

Use cases

  • Bilingual meetings where two people speak different languages and need real-time translation in both directions.
  • Any desktop application with microphone input (Zoom, Teams, Discord, Slack, games, OBS) when using the desktop app.
  • Web-based meeting platforms (Google Meet, Teams, Zoom, Yandex Telemost, Discord, Slack, Gather.town, Whereby, Jitsi Meet) when using the browser extension.
  • Screen sharing live subtitles so the other side can read along.

Getting started

See the repository README for installation instructions. To build from source: bash git clone https://github.com/kizuna-ai-lab/sokuji.git cd sokuji && npm install npm run electron:dev

License and pricing

Licensed under AGPL-3.0. Open source and free to use.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/kizuna-ai-labsokuji.svg)](https://yunodb.com/tools/kizuna-ai-labsokuji?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/kizuna-ai-labsokuji?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/kizuna-ai-labsokuji.svg" alt="Featured on YunoDB" height="20"></a>