hooVA

Inspiration

As the elderly population grows, many individuals face isolation, difficulty using modern technology, and limited access to personal assistance. We were inspired by how AI could bridge this digital divide and so that seniors stay connected, organized, and supported—especially those with memory loss or cognitive decline.

What it does

hooVA is an AI-powered virtual assistant designed specifically for elderly users. It provides an intuitive, voice-enabled interface that allows users to naturally communicate, set reminders, manage daily tasks, and connect with loved ones. By combining speech recognition and intelligent context handling, hooVA helps seniors stay on track without needing to navigate complex technology.

How we built it

We developed hooVA using a Next.js frontend and an Express.js/Node.js backend. We integrated the Gemini LLM for generating context-aware, conversational responses, elevenlabs API for accurate and natural speech-to-text and text-to-speech capabilities, supabase for secure authentication and database management using OAuth.

Challenges we ran into

One of the biggest challenges was fine-tuning the language model. We had the model hallucinate quite a bit, so we had to do a lot of specific prompt engineering.

Accomplishments that we're proud of

We’re proud of creating a fully functional prototype that allows the elderly to interact with technology naturally using their voice. We were also using multiple AI-based APIs, which was an enriching learning experience.

What we learned

We learned how to effectively integrate large language models into real-world applications, manage API pipelines for multimodal interaction, and design with accessibility in mind. We also gained a deeper appreciation for user experience challenges faced by older adults.