Skip to content

A robot that identifies a user's facial expressions and returns Spotify song recommendations based on their emotions.

Notifications You must be signed in to change notification settings

k22jung/emotion_ai

Repository files navigation

Emotion AI

Emotion AI is a small project I completed that identifies a user's facial expressions and emotions and returns song recommendations based on these emotions. The hardware setup consists of a Raspberry Pi connected to a Pi camera mounted to a camera gimbal of two servos. Microsoft Azure's Cognitive Services is used detect facial emotions in images captured by the camera. When image classification data is retrieved by the Pi, the program will create tags associated with the user's current mood and search Spotify for an appropriate song from different playlists. The Pi camera will also detect, lock-on, and follow the user's face using OpenCV and PID control for the gimbal.

Dependencies

Python Packages:

  • numPy
  • json
  • spotipy

Running

Hardware Requirements

You will need to obtain a servo pan & tilt and servos which I found here. You will also need a Pi camera and an external 4 AA battery pack.

Software Requirements

Obtain required Microsoft Azure and Spotify keys and run main.py.

About

A robot that identifies a user's facial expressions and returns Spotify song recommendations based on their emotions.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages