Inspiration
Our group really wanted to do something with artificial intelligence. We all also had a passion for music so our group came to the conclusion to make this project.
What it does
Moodify takes in the users input for mood and recommends a song based on that mood. The user can also input a specific artist they like as well as the mood and that will give a bias in recommending songs from that artist that fit that mood.
How we built it
First we gathered multiple playlists which each playlist is based around a specific mood. From each playlist we then extracted data from each song using python and Spotify API in order get data such as their danceability, loudness, and etc. Then we used a k nearest neighbor algorithm to make classifications about the data of songs of certain moods. We then created a website using HTML and CSS that takes the users input of the artist and the mood of music they want to listen too. The website uses the k nearest algorithm and the user input to recommend songs to them.
Challenges we ran into
The first challenge we ran into was trying to figure out how to use AI in order to figure out how to classify songs. Thanks to one of the mentors, he was able to guide us and teach us about the k nearest neighbor algorithm. Another challenge we faced was trying to use HTML and CSS to create a website considering none of us have much experience with it. However we were able to overcome it with a lot of time and effort.
Accomplishments that we're proud of
We are proud of making separate parts that interwork with each other. Also trying out new things that none of us are very familiar with at all.
What we learned
We learned how to use git to have multiple individuals work on a project at the same time. Also about the k nearest algorithm and how to apply it to a project.
What's next for Moodify
Next is to find a way to have users sign into their Spotify accounts to allow Moodfiy to access their data such as their top played tracks. Once we have the data of their top played tracks, Moodify would be able to recommend songs similar to their top played tracks that fit the mood they have selected.
Log in or sign up for Devpost to join the conversation.