← back
Real-Time Mood Music Recommender

Real-Time Mood Music Recommender

Pending
💰 INR 12500–37500 👤 Unknown 🕒 7d ago status: new
JavaScript Software Architecture Node.js AngularJS Backend Development Frontend Development Computer Vision Web Application
I’m building Face2Tune, a web app that watches a user’s webcam feed, recognises when they look Happy, and instantly queues up music that matches that vibe. The front-end is React; the back-end is a small Node.js / Express.js API. Computer-vision runs in the browser so latency stays low, and I’ll store nothing on the server except the track catalogue. Here’s what I need from you: • Integrate a reliable browser-side emotion-detection model (only the Happy class is required for now, but the code should be easy to extend for Sad or Angry later). • When “Happy” is detected, fetch a curated playlist from the back-end and display it in the UI. • Add a simple toggle that lets the user manually choose their mood instead of relying on the camera. • Keep the React component tree clean, stateful logic in hooks, and provide clear comments so I can iterate on the look and feel. • Supply concise setup instructions (npm install, .env variables, build & run) and a short README explaining how to retrain or swap the model if I decide to expand the emotion set. If the camera feed drops, or the model is uncertain, fall back gracefully to the manual mood picker and surface an unobtrusive toast message; no console errors should slip through linting. Once everything passes those checks, I’ll consider the job complete.
↗ View on Freelancer