Momentune

Rediscover old friendships through shared photos. Browse memories together, reconnect, and forge new bonds—one photo at a time.

Comment

Inspiration

In a world so focused on constant change and innovation, it is easy to get caught up in all the noise and lose sight of the precious memories of old friendships and simpler times. What we love about photo albums is that they invite us to slow down and revisit these cherished memories. Photo albums remind us of the good old days, and more importantly, they remind us of the people we may have lost touch with over time. As such, we decided to create an application to invite people to slow down and go down memory lane together, to rekindle friendships, one photo at a time.

What it does

Momentune is a web app that invites individuals who have drifted apart to reconnect by providing a curated photo album of precious memories they share for them to reminisce. Along with the photo album, Momentune provide a playlist based on the overall emotion of the photo album to further facilitate the feeling of nostalgia.

How we built it

We build the front-end UI using NextJS and React. For the back end, we used FastAPI and Python. To handle user authentication and our database needs, we used Supabase. We used the OpenAI API to analyze the general emotions of each photo album, which was then used to get songs using the Spotify API to create the playlists. The face-recognition Python library was used to recognize old friends in your photos.

Challenges we ran into

One of the challenges we ran into was the implementation of the Spotify API. We had some difficulties dealing with Spotify's API authentication and integrating the API into our application.

Accomplishments that we're proud of

We are proud of being able to find songs based on the general emotion portrayed by a series of images.

What we learned

  • Learned to use Spotify API
  • Learned to use GPT4-V
  • Learned to work with the face-recognition Python library

What's next for Momentune

  • Preparing Momentune for deployment for a more general audience
  • Including more personalization in the curation of playlists
  • Improve the sentiment analysis of images to include a wider range of moods and songs

Built With

Share this project:

Updates