Spotify Mood AR is an experimental augmented reality web app that recommends music based on real-time facial emotion recognition. Using a webcam, the app detects the user's emotional expression and suggests a Spotify track that aligns with their current mood. Users can even change the recommended song by covering their face, triggering a new emotion detection cycle.

My role

I contributed to this project by exploring the intersection of machine learning, computer vision, and music APIs. I worked with pre-trained emotion detection models, integrated the system into a Flask web app, and designed the interface using Figma to ensure a simple, responsive experience. I also helped implement interaction with the Spotify API for music recommendations.

The result

The result is a functional prototype that showcases the potential of combining AI-driven emotion analysis with music personalization. While still in the experimental stage, the app demonstrates innovative use of AR and provides a foundation for future development into real-world, emotion-based music applications.