Limitations of Emotional Expression in AI-Generated Popular Music Creation

Abstract

The usage of artificial intelligence (AI)-generated music technology in popular music production has gained higher popularity. It has been noted though that there is a difference in the intensity of emotional conveying of music generated by AI as compared to the human creation of music. The human beings express the nuanced moods and sentiments in the context of popular music creation through various elements in which human beings strive to create the song structure in ways that are conducive to their tastes, the layers of texture, and the choice of various timbers, thus carrying out the deep meanings of the popular music. Conversely, music AI-generated does not do so in most cases. The paper is dedicated to the constraints of emotional expression in AI-generated popular music and adopts the research method based on the combination of musicological theory and analysis to disclose the problems of the lack of emotional expression methods, low adaptability to context, and inability to adjust cultural element in AI-generated popular music. This study intends to offer musicological theoretical explanation of maximization of the emotional expression of the AI popular music generation technology and encourages improvement of the humanistic aspects of the artificial intelligence in the domain of music production.