MoodMuse

Inspiration

The idea for MoodTune came from our shared experiences of how music has the power to uplift, comfort, or energize us in ways nothing else can. Yet, finding the perfect song or playlist to match how we’re feeling in the moment is often a challenge. We wanted to create a solution that bridges the gap between emotions and music, using the power of AI to make the process seamless and deeply personal. As first-time hackathon participants, we saw this as an opportunity to explore the intersection of technology and creativity, while building something that could genuinely improve people’s daily lives.

What it does

The user logs in with their Spotify account and upon successful login they are sent to our emotion-based music recommender that uses OpenCV to read your current emotion and then recommend you a playlist based off of how you're currently feeling.

How we built it

We started off with installing the dependencies and getting the environment working for all team members. Then, we implemented a face detection programming using ChatGPT prompt programming. Then, we slowly built upon each feature, exploring new libraries like deepface and tensorflow to implement emotion detection. After, we created buttons and passed json requests and responses between the frontend and backend to communicate emotions detected and to return playlists. Then with react, we created a spotify preview and play on spotify button so users can save playlists to their account.

Challenges we ran into

Lots of challenges. A few hours for individual environment setup, facing errors with python version incompatibilities, long loading times for machine learning packages, low frame rate for video preview, and frontend design bugs.

Accomplishments that we're proud of

All of us decided to work with Python for the first time. We explored new packages, features, and actually got a working demo-able product that we're all proud of.