When we got to Hack the North 2019, we had no idea on what we would make. We continued brainstorming until dawn broke at seven in the morning. One of our group members, Swarup began decorating his laptop with stickers. Eventually, he began sticking on the Facebook emoji stickers and that's when we started discussing about how emotions affect the individual. We realized the profound effect that mood has on behavior, and that's when we came to the conclusion that communicating with others in a similar mood increases morale. This gives users the ability to connect with one another at a deeper level.

What it does

This project takes your picture, then uses AI to track your mood and matches you up anonymously with another user that has the same mood. After you are matched up, you are able to chat with them. This helps improve the well-being of our users by always giving them someone they could talk to, without the need to be judged.

How we built it

The website was built using React. We decided to also use Firebase for our login authentication and database because we wanted to make a messaging chat that synced in real time. Finally, we utilized Azure's AI to recognize human emotions from photos.

Challenges we ran into

The main challenge we ran into was extracting the data from the laptop webcam to pass into Azure AI to read the human emotion. This was due to the laptop webcam returning raw data when the Azure AI needed a URL. We struggled for hours looking for a solution to turn the local image to a URL, experimenting with countless amounts of APIs to achieve it.

Accomplishments that we're proud of

We are proud of how much effort we put into this project and the amount of information that we learned in this short period of time.

What we learned

We were all inexperienced with React, but we were able to learn a lot about React after attending the workshop and spending hours learning it. We also learned how to utilize Firebase and Azure APIs.

What's next for Vibe

We plan on adding a questionnaire for the users to answer to better match them with other users.

Share this project: