Chatbot: Music Recommendation System Using Emotion Detection

In the age of personalized entertainment, traditional music recommendation systems often overlook the emotional context of the user. This paper presents an AI-powered chatbot-based music recommendation system that suggests songs dynamically based on the user’s emotional state. The system utilizes Natural Language Processing (NLP), sentiment analysis, and optionally facial and voice inputs to detect emotions such as happiness, sadness, or relaxation. A machine learning model classifies the emotional state, and suitable songs are retrieved from a music database accordingly. The chatbot allows user interaction for refining preferences, including language and artist selection. Built using Python, Flask, and Firebase Firestore, the system supports real-time feedback to improve recommendation accuracy over time. This approach enhances music discovery by aligning songs with the user’s current mood and promotes emotionally adaptive AI. The system demonstrates potential applications in entertainment and wellness, with scope for future upgrades such as deep learning-based emotion detection and multi-modal recommendation strategies for immersive user experiences.