See the AiRA framework in action across different deployment scenarios
Experience the full capabilities of the AiRA receptionist framework through these recorded demonstrations.
Complete walkthrough of the AiRA receptionist interface including chat, voice, and avatar responses.
Demonstration of speech recognition, text-to-speech, and synchronized avatar animation.
YOLOv8-powered visitor presence detection with automatic greeting triggers.
Every demonstration video showcases the core technologies powering the AiRA receptionist experience.
Conversational AI with RAG knowledge retrieval for accurate, context-aware responses tailored to your organization.
Real-time voice input and transcription enabling natural, hands-free interaction with the receptionist.
Text-to-speech with dual voice engines for lifelike, expressive vocal output across multiple languages.
Live2D avatar with lip-sync animation for an engaging, human-like visual receptionist presence.
Visitor presence detection automatically identifies when someone approaches and initiates interaction.
Remote device pairing mode enables seamless integration with kiosks, tablets, and wall-mounted displays.
The AiRA framework is built on battle-tested technologies engineered for reliability at scale.
| Component | Technology |
|---|---|
| Backend | Flask + Socket.IO |
| AI Model | OpenRouter (LLM) |
| RAG | Local knowledge base retrieval |
| Speech | Minimax TTS + ElevenLabs |
| Avatar | Live2D + VTube Studio |
| Detection | YOLOv8 |
| Frontend | Vanilla JS + Jinja2 Templates |
