{"slug": "building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf", "title": "Building AapdaMitra: An AI Disaster Response Voice Assistant for India with Murf Falcon & LiveKit", "summary": "A developer built AapdaMitra, an AI-powered disaster response voice assistant for India, using Murf Falcon, LiveKit Agents, Google Gemini, and Deepgram. The assistant provides safety guidance, weather information, and human escalation through natural voice conversations, addressing challenges such as poor connectivity and user panic during emergencies. The project was developed for the 10 Days of Voice Agents – VoiceForBharat Edition challenge.", "body_md": "*My journey through the 10 Days of Voice Agents – VoiceForBharat Edition challenge.*\n\nNatural voice interaction can make technology more accessible during emergencies. In stressful situations, people often don't have time to type long messages or search through websites. They simply need clear guidance.\n\nTo solve this problem, I built AapdaMitra, an AI-powered Disaster Response Voice Assistant designed for India.\n\nAapdaMitra helps users during emergencies such as floods, cyclones, earthquakes, fires, landslides, storms, and heatwaves through natural voice conversations.\n\nThe project was built as part of the 10 Days of Voice Agents – VoiceForBharat Edition challenge using Murf Falcon, LiveKit Agents, Google Gemini, and Deepgram\n\nDuring disasters:\n\nPeople may panic.\n\nThey may not be able to type.\n\nInternet connectivity may be poor.\n\nQuick spoken guidance can save valuable time.\n\nVoice is the most natural interface during emergencies.\n\nMeet AapdaMitra\n\nAapdaMitra is a conversational AI assistant that provides:\n\nDisaster safety guidance\n\nWeather information\n\nUser memory\n\nHuman escalation\n\nOutbound voice calls\n\nSpecialist agent handoffs\n\nCall analytics dashboard\n\nThe assistant always prioritizes user safety while avoiding hallucinations through carefully designed guardrails.\n\nFrontend\n\nBackend\n\nAI Models\n\nDatabase\n\nOther\n\n```\nUser Voice\n      │\n      ▼\nDeepgram Speech-to-Text\n      │\n      ▼\nGoogle Gemini\n      │\n      ├── Memory Tool\n      ├── Weather Tool\n      ├── Escalation Tool\n      ├── Analytics\n      └── Specialist Agent\n      │\n      ▼\nMurf Falcon Text-to-Speech\n      │\n      ▼\nLiveKit\n      │\n      ▼\nUser\n```\n\nThis project was far from straightforward.\n\nOne major challenge was designing prompts that made Gemini collect all required user information before triggering tools like human escalation. Initially, the model skipped questions and called tools with incomplete data.\n\nThe solution involved redesigning the workflow so the assistant gathered missing information one question at a time and only invoked the tool after every required field had been collected.\n\nAnother challenge was integrating outbound SIP calling. Configuring LiveKit, Linphone, SIP trunks, and testing outbound calls required careful setup and troubleshooting.\n\nI also encountered issues with participant handling, webhook integration, session management, and UI synchronization before arriving at a stable implementation.\n\nClone the repository.\n\n`git clone <your-repository-url>`\n\nInstall dependencies.\n\n`uv sync`\n\nConfigure environment variables.\n\n```\nLIVEKIT_URL=\nLIVEKIT_API_KEY=\nLIVEKIT_API_SECRET=\nGOOGLE_API_KEY=\nDEEPGRAM_API_KEY=\nMURF_API_KEY=\n```\n\nRun the backend.\n\n`uv run python src/agent.py dev`\n\nStart the frontend.\n\n```\nnpm install\nnpm run dev\n```\n\nOpen the browser and begin a voice conversation.\n\nThe 10 Days of Voice Agents – VoiceForBharat Edition challenge provided an excellent opportunity to explore real-time conversational AI.\n\nBuilding AapdaMitra helped me understand how voice, AI, and real-time communication can work together to create technology that has the potential to make a meaningful impact during emergencies.\n\nA big thank you to Murf AI for organizing this challenge and providing the opportunity to build with Murf Falcon, along with the powerful ecosystem around LiveKit.", "url": "https://wpnews.pro/news/building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf", "canonical_source": "https://dev.to/nayabshah/building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf-falcon-livekit-2d9n", "published_at": "2026-08-15 16:23:03+00:00", "updated_at": "2026-08-15 16:42:33.089119+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-products", "ai-tools", "natural-language-processing", "generative-ai"], "entities": ["AapdaMitra", "Murf Falcon", "LiveKit", "Google Gemini", "Deepgram", "Murf AI", "VoiceForBharat"], "alternates": {"html": "https://wpnews.pro/news/building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf", "markdown": "https://wpnews.pro/news/building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf.md", "text": "https://wpnews.pro/news/building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf.txt", "jsonld": "https://wpnews.pro/news/building-aapdamitra-an-ai-disaster-response-voice-assistant-for-india-with-murf.jsonld"}}