Today, a video released by MiniMax, a Shanghai-based general artificial intelligence company, has garnered widespread online attention. The video's visuals were generated by the "Conch" video model, while the music was generated by the Music 1.5 model. The moving vocals of a black singer, the melodious notes of a saxophone, the passionate drumming of a drum set... The music generated by Music 1.5 is highly captivating, and at four minutes long, it represents a significant improvement over the previous generation of music generation models. The MiniMax R&D team explained that this new generation of music generation models has achieved several technological breakthroughs. In simple mode, users simply input a few keywords or a natural language description, and it generates a highly complete song. In advanced mode, users can customize the music's features using "16 styles, 11 emotions, and 10 scenes," creating lyrics for different sections of the song, such as the intro, verse, and chorus, enriching the user's creative experience.