All of us have seen music shape our lives in different ways, and we're obsessed with constantly and meticulously updating our playlists. Because of this, we simply wanted to find an easier way to capture a mood with a playlist. Thus, we decided to develop Thesia.

What it does

Thesia generates a playlist based off of an uploaded image. We use Google Cloud Platform's Vision API to identify key labels associated with every image, and use these labels, along with our mood identification algorithm (based on the color properties of the image), and used these, integrated with Spotify's API, to generate a playlist that best matched the mood of the image (using variables such as danceability, acousticness, and more).

How we built it

We used Node.js and a combination of HTML/CSS/JavaScript for the front end, and used both the Google Cloud Platform (specifically the Vision API) and the Spotify Developer API.

Challenges we ran into

We were unable to completely debug our program, so while all of the parts work individually, they run into problems when integrated within each other.

Accomplishments that we're proud of

We created an app that we would genuinely use for ourselves and used technologies we had never used before to create it.

What we learned

We learned more about the Google Cloud API and the Spotify API, and all learned how to use Node.js over the course of the weekend.

What's next for Thesia

Hopefully, debugging the software and launching it on it's own domain (and maybe making a little cash on the side ;) )

Share this project: