Inspiration To create a warm, human‑like video companion that empowers seniors with personalized memories, proactive wellness nudges, and seamless family connections.
What it does EchoCare AI enables live video chat powered by Tavus, logs mood, offers health prompts, memory‑lane storytelling, and a “Loved Ones Circle” for audio messages.
How we built it We used React/Framer‑Motion for UI, Supabase for persistence, Sonner for toasts, Tavus API for video & voice, and custom hooks to manage camera, transcripts, and proactive prompts.
Challenges we ran into Integrating real‑time Tavus conversations, managing iframe postMessage for speech, and ensuring async prompts felt natural rather than robotic.
Accomplishments that we're proud of Seamless long‑term memory logging, natural‑feeling proactive nudges, and bi‑directional loved‑ones audio messaging within one cohesive interface.
What we learned Balancing proactive AI engagement with user comfort is key; async flows and timed prompts must be carefully tuned to feel supportive, not intrusive.
What’s next for EchoCare Enhance emotion detection via video & voice, add group family calls, refine scheduling for daily check‑ins, and integrate wearable data for smarter health nudges.
Built With
- elevenlabs
- react
- tailwind
- tavus
Log in or sign up for Devpost to join the conversation.