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.

Automatically checked · GitHub checked 2026-11-18

sokuji

Overview

Sokuji Logo

Real-time two-way speech translation for bilingual meetings — cloud or fully offline on your device

AGPL-3.0 License Build and Release Latest Release Platforms Ask DeepWiki

English | 日本語 | 中文 | Tiếng Việt


Why Sokuji?

Built by Kizuna AI Lab — we use AI to break language and accessibility barriers, creating genuine human connections. "Kizuna" (絆) means "bond" in Japanese, and Sokuji (即時) is our flagship tool to make real-time communication possible across any language.

Sokuji is a cross-platform live speech translation app for desktop and browser. It supports Local Inference — on-device ASR, translation, and TTS powered by WASM and WebGPU, with no API key required, no expensive GPU needed, fully offline, and completely private. It also integrates with cloud providers including OpenAI, Google Gemini, Palabra.ai, Kizuna AI, Doubao AST 2.0, Soniox, and OpenAI-compatible APIs.


How It Works

graph LR
    A["🗣️ You speak<br/>(any language)"] --> B["🎙️ Sokuji"]
    B --> C{"Choose one"}
    C -->|"☁️ Cloud"| D["OpenAI · Gemini<br/>Palabra · Doubao..."]
    C -->|"🖥️ Local"| E["On-device AI<br/>ASR → Translate → TTS<br/>(fully offline, no GPU)"]
    D --> F["🔊 Translated voice"]
    E --> F
    F --> G["💻 Zoom · Teams · Meet<br/>Discord · Any app"]

    style A fill:#4a9eff,stroke:#357abd,color:#fff
    style B fill:#10a37f,stroke:#0d8a6a,color:#fff
    style C fill:#ff9f43,stroke:#e88a2e,color:#fff
    style D fill:#6c5ce7,stroke:#5a4bd1,color:#fff
    style E fill:#00b894,stroke:#009d7e,color:#fff
    style F fill:#fd79a8,stroke:#e56b96,color:#fff
    style G fill:#636e72,stroke:#525c60,color:#fff
Providers8 — OpenAI, Gemini, Palabra.ai, Kizuna AI, Doubao AST 2.0, Soniox, OpenAI Compatible, Local Inference
Local Models44 ASR models, 75 translation models, 137 TTS models
Languages99+ (speech recognition) · 55+ (translation) · 53 (text-to-speech)
PlatformsLinux · Windows · macOS · Chrome · Edge
PrivacyLocal Inference = 100% on-device, no API key, no internet

Demo

Open video on GitHub

Install

Sokuji is available as a Desktop App and a Browser Extension — same features, different reach.

Desktop AppBrowser Extension
FeaturesAll features identicalAll features identical
Use withAny app with mic input — Zoom, Teams, Discord, Slack, games, OBS, and moreWeb-based meeting platforms — Google Meet, Teams, Zoom, Yandex Telemost, Discord, Slack, Gather.town, Whereby, Jitsi Meet
InstallDownload & installZero install — add from store
PlatformsWindows · macOS · LinuxChrome · Edge · Brave (coming soon)

Desktop App

Download from the Releases page:

PlatformPackage
WindowsSokuji-x.y.z.Setup.exe
macOS (Apple Silicon)Sokuji-x.y.z-arm64.pkg
macOS (Intel)Sokuji-x.y.z-x64.pkg
Linux (Debian/Ubuntu x64)sokuji_x.y.z_amd64.deb
Linux (Debian/Ubuntu ARM64)sokuji_x.y.z_arm64.deb

Browser Extension

Available on Chrome Web Store Available on Microsoft Edge Add-ons

Install extension in Developer Mode
  1. Download sokuji-extension.zip from the Releases page
  2. Extract the zip file
  3. Go to chrome://extensions/ and enable "Developer mode"
  4. Click "Load unpacked" and select the extracted folder

Build from Source

git clone https://github.com/kizuna-ai-lab/sokuji.git
cd sokuji && npm install
npm run electron:dev        # Development
npm run electron:build      # Production

Two-Way Translation for Bilingual Meetings

Two people, two languages, one conversation. Sokuji auto-detects which language is being spoken and translates it into the other — in both directions, in real time. Set Language A and Language B, capture system audio and microphone together, and everyone follows along.

  • Auto language detection — no manual switching mid-conversation
  • Both directions at once — mic + system audio in a single session
  • Live subtitles — share your screen so the other side reads along too
  • Works in Zoom, Google Meet, Teams, Discord, or any app (desktop), and directly inside web meeting platforms (browser extension)

Powered by Soniox two-way mode — 60+ languages, 3,600+ language pairs.


Features

Local Inference (Edge AI)

Run everything on your device — no API keys, no internet once models are downloaded, no expensive GPU, complete privacy. Powered by WASM and WebGPU, Sokuji runs efficiently on any modern browser using your existing CPU and integrated graphics.

  • 44 ASR models (23 offline + 10 streaming + 11 WebGPU including Whisper, Cohere Transcribe, Voxtral, Granite Speech) covering 99+ languages
  • 75 translation models — 69 Opus-MT language pairs + 6 multilingual LLMs (Qwen 2.5 / 3 / 3.5, Hunyuan-MT 1.5, TranslateGemma) with WebGPU
  • 137 TTS models across 53 languages (Piper, Piper-Plus, Coqui, Mimic3, Matcha, MMS, VITS, Supertonic engines)
  • One-click model download with IndexedDB caching

Cloud Providers

ProviderKey Feature
OpenAIgpt-realtime-mini / gpt-realtime-1.5 · 10 voices · configurable turn detection (Normal / Semantic / Disabled) · noise reduction · 60+ languages
Google GeminiDynamic model selection (audio/live models) · 30 voices · built-in turn detection · 34 language variants
Palabra.aiWebRTC low-latency · voice cloning · auto sentence segmentation · partial transcription translation · 60+ source / 40+ target languages
Kizuna AISign in and go — API key managed by backend · same OpenAI models with optimized defaults
Doubao AST 2.0Speech-to-speech with speaker voice cloning · bidirectional Chinese↔English · Ogg Opus audio output
SonioxReal-time speech-to-speech · one-way & two-way translation (auto-detects speaker language) · 60+ languages / 3,600+ pairs · 200 voices (filter by gender, age, accent, use case, style) · voice cloning · bring-your-own-key
OpenAI CompatibleBring your own endpoint — any OpenAI Realtime API-compatible service (Electron only)
Local InferenceFully offline · ASR → Translation → TTS on-device · no API key · no GPU required

Audio

  • Translate your voice — speak in your language, others hear the translation as if you spoke it natively
  • Translate others' voice — capture meeting audio (extension) or any system audio (desktop) and get real-time translated subtitles
  • Virtual Microphone — route translated audio to Zoom, Meet, Teams, or any app
  • Real-time Passthrough — monitor your own voice while recording
  • AI Noise Suppression — removes background noise, keyboard sounds, and other distractions
  • Echo Cancellation — built-in with modern Web Audio API

Interface

  • 30 languages — fully localized UI
  • Simple Mode — streamlined setup for non-technical users
  • Advanced Mode — waveform display and detailed controls

Privacy

Your audio stays on your device — if you choose Local Inference, your audio and text never leave.

  • Cloud mode connects directly to provider APIs — no intermediary servers
  • API keys stored locally only, never transmitted to us
  • Local Inference processes audio and text on-device; anonymous usage analytics are still sent to PostHog
  • Anonymous usage analytics via PostHog

Tech Stack


Contributing

We welcome contributions! Please read our Contributing Guidelines before getting started.


License

AGPL-3.0

Sponsors

SignPathFree code signing on Windows provided by SignPath.io, certificate by SignPath Foundation.

Support

Acknowledgments

For detailed model licenses, see THIRD_PARTY_NOTICES.md.