Real-time multilingual conversation platform powered by Azure Cognitive Services. Speak in one language, hear and read translations in another — instantly. Share sessions with guests via QR code so anyone can follow along in their preferred language.
Continuous transcription powered by Azure Speech Services. Supports automatic language detection from 50+ languages or manual language selection. Audio streams directly from the browser microphone — no plugins or extensions required.
Instant translation between 10+ supported languages via Azure Translator. Choose between two modes:
- Standard mode — Sentence-level translation with higher accuracy
- Real-time mode — Streaming translation with lower latency, words appear as you speak
Tap any utterance to hear it spoken aloud in the translated language. Uses Azure Neural Voices for natural-sounding speech synthesis.
Invite anyone to follow along — no account required. Guests use their own device's microphone to capture audio locally, so each participant gets the best possible input from their own hardware rather than relying on a single shared mic across the room.
- Click Invite to generate a shareable link with QR code
- Guests scan the QR code or open the link on any device — phone, tablet, or laptop
- Guests choose their preferred language and speak into their own microphone
- Everyone sees live translations in real time, powered by each device's local audio
Every conversation is automatically saved with full transcript replay.
- Browse past sessions by date, duration, and participant count
- Replay transcripts with synchronized audio playback
- Export conversations as JSON or formatted text
Color-coded speakers with customizable labels make it easy to follow who said what in multi-participant conversations.
Record raw microphone audio alongside transcripts. Audio chunks are stored in Azure Blob Storage and linked to session history for later playback.
- Sign in with your Microsoft account
- Select your languages (e.g., English ↔ Arabic)
- Click Start Listening to begin real-time transcription
- Speak naturally — translations appear instantly in both language panels
- Navigate to the History tab
- Browse past sessions with dates, durations, and participant counts
- Click any session to replay the full transcript
- Play back recorded audio alongside the transcript
- Microphone selection — Choose your preferred input device
- Translation mode — Switch between Standard and Real-time translation
- Detection mode — Auto-detect spoken language or manually specify
git clone https://github.com/your-org/live-translation.git
cd live-translation
npm install && cd api && npm install && cd ..
cp .env.example .env # Fill in your Azure resource values
npm run dev # https://localhost:5173See the Development Guide for full setup instructions.
| Document | Description |
|---|---|
| Architecture | System overview, component details, security model, project structure |
| Development | Local setup, scripts, testing, Docker |
| Infrastructure | Terraform provisioning, Azure resources, variables, cost estimate |
| Deployment | Frontend & API deployment, CI/CD setup, post-deploy checklist |
MIT — see LICENSE.