DJ-AI: Intelligent Playlist Sorting and Seamless Generative Transitions

The music listening experience has improved and become more accessible over time with digital playlists and real-time DJ transitions. However, existing systems do not cover fully amateur DJs and end users. Unlike traditional recommendation algorithms, DJ-AI analyzes songs not only based on user preferences but also their musical features to generate optimal sequences and seamless transitions. Graph-based optimization methods have been used for playlist arrangement, while MusicGen and MERT embeddings have been integrated to enhance transition smoothness. Experimental results show that the proposed method improves the listener experience by increasing transition compatibility.