Emotion-Based Music Suggestor - README Project Overview The Emotion-Based Music Suggestor is a Python-based application that captures a picture of your face using your webcam, detects your emotional state using the DeepFace module, and recommends YouTube music videos that match your current emotion. The project leverages computer vision (OpenCV) for face detection and the DeepFace library for emotion analysis. Once the emotion is identified, the application generates a list of music videos from YouTube that corresponds to the detected emotion.
Features Capture a live image using a webcam via OpenCV. Detect facial emotions using the DeepFace module. Suggest relevant YouTube music videos based on the detected emotion. A simple, interactive UI/landing page displaying music suggestions. Requirements Before you start, ensure you have the following software and libraries installed:
Prerequisites Python 3.x OpenCV (cv2) DeepFace Flask (for the web interface) Requests (for interacting with the YouTube API) YouTube Data API (for fetching relevant YouTube video links) Other dependencies mentioned in requirements.txt Customization You can customize this project by:
Adjusting the music suggestions for each emotion. Adding more functionality, like personalized playlists based on user preferences. Integrating Spotify or other music platforms. Troubleshooting Webcam Issues:
Ensure your webcam is functional. If OpenCV cannot access your webcam, ensure that no other application is using it and that your browser has given the necessary permissions. DeepFace Errors:
Ensure you have installed all dependencies of DeepFace (tensorflow, keras, etc.). If emotion detection fails, check the captured image quality. Poor lighting may affect results. Future Enhancements Improved Emotion Detection: Enhance the emotion detection algorithm by training it on a more diverse dataset for improved accuracy. Music Platform Integration: Extend support to Spotify or Apple Music for personalized music recommendations. Mobile Version: Develop a mobile-friendly version of the app. License This project is licensed under the MIT License.
Happy coding! Enjoy your emotion-based music recommendations! 🎵