The Borderless Collaboration Workspace for Global Teams.
LingoLive is a real-time collaboration platform that eliminates language barriers in remote work. It combines instant chat translation, voice synthesis, and developer tools into a single, seamless workspace.
Built for the Lingo.dev Hackathon.
(Click the image to watch the demo)
- Zero-Latency Experience: Messages appear instantly (Optimistic UI) while translation happens in the background.
- Context-Aware: Uses the Lingo Engine to understand context (e.g., distinguishing "Bank" as a financial institution vs. a river bank).
- Cross-Language Support: Users see messages in their preferred language (English, Spanish, French, German, Japanese, Nepali).
- "Listen" to any Message: Integrated browser-native Speech Synthesis.
- Language Auto-Detection: Automatically selects the correct voice pack (e.g.,
ja-JPfor Japanese) to ensure accurate pronunciation. - Accessibility: Makes global communication accessible to everyone.
- JSON Localization Automator: Solves the pain of manual software localization.
- Visual Split-View: Compare original
en.jsonvs translatedes.jsonside-by-side. - Recursive Parsing: Smartly handles nested JSON objects preventing crashes on complex config files.
- One-Click Download: Get production-ready localized files in seconds.
- Intelligent Code Scanning: A custom script (
audit-cli.js) that scans the codebase for hardcoded strings and brand violations. - CI/CD Integration: Runs automatically in GitHub Actions to prevent non-localized code from hitting production.
- MCP Architecture: Built with Model Context Protocol principles for future AI agent integration.
- Framework: Next.js 14+ (App Router)
- Styling: Tailwind CSS + Framer Motion (Glassmorphism Design)
- Backend/Realtime: Supabase (PostgreSQL + Realtime Channels)
- Lingo.dev Ecosystem:
- SDK:
lingo.dev/sdk(Real-time Context & Translation) - CLI: Custom
audit-cli.js(Built with Node.js) - MCP: Model Context Protocol Server (
scripts/mcp-server.js) - GitHub Actions: Automated CI pipeline (
.github/workflows/lingo.yml)
- SDK:
- CI/CD: GitHub Actions (Automated Audits)
git clone https://github.com/Rupe88/lingolive-hackathon.git
cd lingolive-hackathonnpm installCreate a .env.local file in the root directory:
NEXT_PUBLIC_LINGO_API_KEY=your_lingo_key
NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_keynpm run devOpen http://localhost:3000 in your browser.
- Login: Enter your name and select a language (e.g., "Japanese").
- Chat: Type a message in English. See it appear in Japanese for the receiver. Click the "Speaker" icon to hear it.
- File Hub:
- Go to the File Hub tab.
- Drag & Drop the provided
demo-assets/en.jsonfile. - Click Start Localization.
- See the real-time split view comparison and download the result.
During testing, 8 of my college friends joined the same LingoLive room. We spent hours chatting, shouting random words, and testing the limits of the translation engine (even with some colorful language 😂). It proved that language barriers can't stop a good time!
MIT License. Created by Rupesh for the Lingo.dev Hackathon 2026.

