MoodTunes

• 

  Camera

• 

  Text/Voice

• 

  Vibe

• 

  Favorites

Inspiration:

Our inspiration came from the frustration of trying to create playlists based on how we were feeling or what we were doing in the moment. It was always hard to translate emotions into search terms, you can’t just type a feeling into a music search bar and get the perfect songs. We wanted to solve that by building a platform where you can simply express your mood and instantly get a personalized playlist that matches it.

What it does:

MoodTunes is a music recommendation platform that creates personalized playlists based on how you’re feeling in the moment. Users can type their mood, speak it, or use a camera that detects facial expressions to understand their emotion. Based on this input, MoodTunes suggests songs that match the user’s vibe. It also allows users to explore new music and save their favorite tracks for later.

How we built it:

We built MoodTunes as a full-stack application using a React frontend and a Python (FastAPI) backend. The frontend handles user interaction, including mood input, voice input, and webcam-based emotion detection, while the backend processes the input and returns song recommendations. We integrated emotion detection models to interpret facial expressions and map them to moods. Claude helped us throughout the development process by assisting with debugging, structuring components, and improving the overall logic and design of the system.

What's next for MoodTunes:

Next, we want to improve the frontend design to make MoodTunes more intuitive, modern, and visually engaging. We also plan to enhance our AI model so it can better understand emotions and provide more accurate, real-time music recommendations. In addition, we aim to expand features based on user feedback, so the platform continues to evolve around what users actually want and need.

Built With

• css
• face-api.js
• fastapi
• huggingface
• javascript
• last.fm-api
• python
• pytorch
• react
• uvicorn