POINT | Hackathon Winner | Most Creative Track | TechNYU

#hackathon #texttospeech #ai #webdev

I recently participated in a Hackathon, organized by TechNYU. I had an amazing team—Brandon, Russel, and Faizan—who brought their expertise and creativity to the table.

Together, we developed POINT (Point Of Interest Narrating Tourguide), a web application designed to revolutionize the way people explore new places. Our hard work paid off, as we won the Most Creative Track with POINT!

Inspiration

When people travel, they often miss out on fascinating landmarks or local points of interest, or they simply wish they’d known more about them beforehand. Our app fixes that by tracking every noteworthy spot along your route and providing you with a deep dive into its history. Whether it’s a well-known sight or a hidden gem, we’ll make sure you’re fully immersed in the story behind it—so you can get the most out of your journey, every step of the way.

What it does

POINT is a web application that allows users to enter a starting point and a destination, generating a route with key points of interest (POIs) along the way. Users can listen to AI-generated audio descriptions of these locations, making their journey more engaging and informative without needing a tour guide.

How we built it

Fully Unit, E2E, and Integration Tested
CI/CD github workflow actions
Frontend: Built using Next.js with TypeScript for a seamless and responsive user experience, styled with Tailwind CSS.
Backend: A Flask-based API that fetches location descriptions using Claude AI and converts text to speech using ElevenLabs.
Google Maps Integration: to display routes and POIs interactively.
State Management: Implemented using React hooks for smooth data flow.

Challenges we ran into

API Integration: Ensuring smooth communication between the frontend and backend, especially handling audio file generation and streaming efficiently.
Optimizing Route Suggestions: Balancing accuracy and performance while displaying POIs dynamically along a route.
Scalability: Designing the system to handle increasing numbers of users and location requests without delays.

Accomplishments that we're proud of

Successfully integrating AI-powered location descriptions and real-time text-to-speech conversion.
Creating an intuitive and visually engaging UI that enhances the user experience.
Implementing a responsive and scalable backend that efficiently serves audio descriptions for various locations.

What we learned

The importance of optimizing API calls for performance and cost-effectiveness.
How to improve user experience by combining interactive maps with real-time audio narration.
The challenges and solutions involved in integrating multiple third-party APIs in a seamless manner.