A mood-aware music app. It reads your emotion from a selfie, your stress from your voice and the weather outside, then plays a playlist that fits and learns how you feel after.

[ Role ]

Flutter developer, ML integration

[ Categories ]
  • Mobile (Flutter & iOS)
  • AI & Automation
[ Stack ]
  • Flutter (Dart)
  • TensorFlow
  • Python
  • FastAPI
  • Firebase
  • Weather API

Music apps recommend from listening history, but ignore how you feel right now. VibraSync starts from your current state. Cam Mix reads emotion from a selfie, Voice Tunes detects mood and stress from a short voice clip, Age | Gender adds a demographic signal, and 360 Mix blends them with the local weather. A teammate built the TensorFlow models; I built the Flutter app, including the Vibra Player, and connected it to the models through FastAPI. After listening, the app asks how you feel and stores it in Firebase.

  1. 01Camera and mic to model. I built the selfie and voice capture flows and sent them to the team's TensorFlow models through FastAPI, turning results into a mood.
  2. 02Four mood modes. I designed the app around Cam Mix, Voice Tunes, Age | Gender and 360 Mix, so users choose how much they share to get a playlist.
  3. 03Vibra Player. I built the in-app player with play, pause, repeat and skip, plus a refresh button that asks the models for a fresh recommendation.
  4. 04Learns from feedback. After a session the app asks how you feel and saves it to Firebase, so the next recommendations get closer to what works.

Want something like this built?