Inspiration

The inspiration for EchoCare came from seeing loved ones struggle with dementia, where forgotten conversations and confusion often lead to isolation. We wanted to build a compassionate, tech-driven solution to help preserve everyday interactions and offer both memory support and peace of mind.

What it does

EchoCare is an AI-powered app that:

  • Transcribes daily conversations into readable text
  • Analyzes them using NLP to extract key information
  • Automatically generates to-do lists and memory cues
  • Supports caregivers with insights, all while keeping data secure and private

How we built it

  • Frontend: React.js for the interface, focusing on accessibility
  • Backend: Node.js with Express to handle API calls
  • AI/NLP: Integrated Google Speech-to-Text and natural language models for summarization
  • Database: MongoDB for secure user data storage
  • Privacy: End-to-end encryption for user trust and HIPAA-compliant practices

Challenges we ran into

  • Handling noisy audio in real-world scenarios
  • Keeping the app lightweight yet functional for older devices
  • Ensuring AI-generated tasks remained contextually accurate
  • Building a UI intuitive for elderly users without tech experience

Accomplishments that we're proud of

  • A fully working prototype that transcribes and summarizes in real time
  • Creating a seamless experience for both patients and caregivers
  • Designing an interface that respects user dignity and simplicity
  • Implementing strong data privacy features from the ground up

What we learned

  • The importance of empathetic design when working with vulnerable users
  • Advanced speech-to-text and NLP integration
  • Best practices for health-tech data privacy
  • How to balance technical performance with emotional impact

What's next for EchoCare

  • Adding voice-based reminders and calendar syncing
  • Expanding to multi-language support for broader accessibility
  • Integrating caregiver dashboards for shared memory tracking
  • Partnering with health orgs to pilot in real-world care settings

Built With

  • express.js
  • jwt
  • mongodb-atlas-(free-tier)
  • n8n
  • node.js
  • openai-(free-tier)
  • react.js
  • tailwind-css
  • vite
  • web-speech-api
Share this project:

Updates

Submission history