Inspiration

I saw students in Devon Hall getting really stressed during long coding sessions. Computers are usually blind to how we feel inside. I wanted to build something that lets your laptop understand your body so it can help you stay calm and focused.

What it does

Bio-Beat Sync is like a co-pilot for your health. It uses your webcam to watch your pulse without touching you. If it sees your heart rate go up while you are working, it uses AI to figure out if you are stressed. Then, it talks to you using a kind voice from ElevenLabs and finds a YouTube Music playlist to help you relax.

How we built it

I used Python for the main logic and OpenCV to watch the face. I wrote math code to find the heart rate by looking at tiny color changes in the skin. I built a login and sign up system that saves your profile to a MongoDB database. For AI, I used Groq because Gemini was giving me errors during the build, but I plan to switch to Gemini in the future.

Challenges we ran into

The biggest wall was the Spotify API being on hold for new apps, so I could not connect it directly. I also faced major issues while integrating ElevenLabs as the program kept crashing during voice generation. Some YouTube video links provided by the AI were unavailable as well. I also worked hard on making the heart rate math stay accurate even when the user moves their head.

Accomplishments that we're proud of

I am proud that I made a "Contactless" heart monitor work just using a normal laptop camera. It was amazing to see the AI recognize a stress spike and immediately try to help. I also built a full database and auth system from scratch in just 24 hours.

What we learned

I learned a lot about signal processing and how to use Fast Fourier Transforms to find a pulse in video. I also learned how to pivot my project when an API does not work. This hackathon taught me how to combine biology with artificial intelligence to solve a real human problem.

What's next for Bio-Beat Sync

I want to fix the ElevenLabs crashing issue to make the voice coach more stable. I also want to make the heart beat detection even more accurate in low light. My big goal is to get a partnership with Spotify so the music plays automatically. I want to turn this into a tool that every student at OU can use to beat burnout.

Built With

Share this project:

Updates