Inspiration
My team was inspired to build this project after we recognized how enjoyable a music listening experience can be if it is tailored to your current emotions.
What it does
This program first uses facial recognition and machine learning to classify a user's emotion. Then, the program provides song recommendations based on both the mood and the user's prior listening history on Spotify.
How we built it
We built this program using the spotipy and deepface libraries and frameworks. The deepface library does the facial recognition and records the user's current emotion. Next, we use this emotion to tailor certain variables of songs like danceability and pass this information to Spotify's API using the spotipy library. From this, we get a list of recommendations which are then outputted to the user via the terminal.
Challenges we ran into
There were many challenges we ran into regarding making requests and getting responses from Spotify's developer API that gives us access to its database.
Accomplishments that we're proud of
Collaborating successfully in a group to make a working prototype for a complex idea
What we learned
We learned how to use the Spotify API and the deepface framework. We also learned a lot of networking requests and parsing their responses.
What's next for MoodMusic
Right now we are just outputting the results to a terminal but an actual user interface would definitely make for a better experience.
Log in or sign up for Devpost to join the conversation.