Music Fun

What is Music Fun :3?

Music Fun :3 is our submission for TartanHacks 2023! Our project features the following functionalities:

• An aesthetic and interactive graph-based display of music search results, complete with album covers and links to the spotify recommendations!
• Song recommendations based on input text emotion!

How we made it

This project was built with a html/css javascript frontend and python backend. We used D3.js to make the amazing interactive graph! Music Fun used the emotify model to classify the emotion of a particular phrases of text. Afterward encoding the classification, a random seed artist and song is chosen based (this is to put into our Spotify API request). A GET request is sent to the spofity recommendation API. After extracting the responses from the API, we associate each recommendation with a node and connect the nodes together to form a graph representation of the results!

What we would have done if we had more time

• Randomly distribute node connections
• Further polish our graph aesthetics
• Allow for variable numbers of requests (we currently only fetch 12 requests each time)
• Add varying more seeds for artists and songs to increase randomization

Short demo of our project in action!

Modules for Development

pip install tensorflow
pip install flask
pip install flask_co
pip install pandas
pip install numpy
pip install scikit

Have fun~~~ :3333