Composite Artificial Intelligence

10Composite AI-Driven Music Recommendation: Integrating Emotion, Aural Analysis and Song Similarity

Ayushmaan Das and Rajalakshmi Shenbaga Moorthy^*

Sri Ramachandra Faculty of Engineering and Technology, Sri Ramachandra Institute of Higher Education and Research, Chennai, India

Abstract

The emergence of digital consumption and streaming services such as Spotify has made them central in the discovery of music, offering users vast collections to choose from. However, it remains challenging to navigate those large collections to serve personalized and emotionally relevant content. The proposed composite AI-driven music recommendation engine fills these gaps by leveraging technologies such as cosine similarity for matching the songs, VGG-16-based CNN for mood-based recommendation, LSTMs for lyric generation, and a Pandas AI chatbot for interactive user engagement, to seamlessly orchestrate an immersive music experience. Current music recommendation systems frequently fail to provide emotionally compelling or contextually relevant suggestions, even with the advancements in recommendation algorithms. More complicated emotional indicators or the subtle aural preferences of users are sometimes overlooked in favor of recommendations that are based on past user behavior or basic genre classifications. Systems that can interpret user data and take into consideration the emotional and aural environments that shape musical choices are becoming more and more necessary. The main goal is to ...