#3535th of 81 in VoiceFor voice, speech and music
sokuji
Live two-way speech translation for meetings, with cloud or on-device models
- Stars
- 1.5k
- License
- AGPL-3.0
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
784 on 21 Apr 20261,487 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Sokuji is an Electron desktop app and Chrome/Edge extension that translates speech in real time and routes the translated voice to Zoom, Meet, Teams or any app through a virtual microphone. It supports eight providers, including OpenAI, Gemini, Soniox and Palabra.ai, plus a local mode that runs ASR, translation and TTS on-device via sherpa-onnx, Transformers.js, WASM and WebGPU. Two-way mode auto-detects the speaker's language through Soniox.
Who it is for: People holding bilingual meetings who want live translation, cloud or offline
Strengths
- Local mode runs ASR, translation and TTS offline with no API key or GPU
- Ships 44 ASR, 75 translation and 137 TTS models, with one-click download
- Virtual microphone and live subtitles work with any meeting app
- Same features in desktop app and Chrome/Edge extension
Weaknesses
- Two-way auto language detection depends on the Soniox cloud provider
- OpenAI-compatible provider is Electron only, not in the extension
- Sends anonymous usage analytics to PostHog
- AGPL-3.0 license restricts closed-source and network-service reuse
What it needs
- GPU optional
- Needs OpenAI, Gemini, Palabra.ai, Soniox or Doubao API (cloud mode only)
- Models: sherpa-onnx, Whisper, Opus-MT, Qwen, TranslateGemma
Quick answers
Is sokuji open source?
Yes. Its code is on GitHub under the AGPL-3.0 license.
Does sokuji need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in Voice
See all 81| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 98 out of 100 |
| 2 | handyOffline push-to-talk dictation app that pastes transcripts into any text field | 91 out of 100 |
| 3 | whisper.cppC/C++ speech-to-text engine running OpenAI Whisper models locally | 90 out of 100 |
| #4 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 83 out of 100 |
| #5 | buzzOffline desktop app for transcribing and translating audio with Whisper | 83 out of 100 |