Studies of Algorithmic Music Generation; Folksong Enthusiasm in Post-Cultural Revolution China

Abstract

This dissertation consists of two independent chapters: "Studies of Algorithmic Music Generation" and “Folksong Enthusiasm in Post-Cultural Revolution China.” The chapter "Studies of Algorithmic Music Generation" investigates algorithms for creating "long-term structure" in music. The chapter starts by defining the concept of "long-term structure" and underscoring its significance in automated music generation through a review of historical literature and musical examples. The chapter then introduces two algorithmic techniques to realize this concept: one called "long-term structure sampling" and another applying the image generation technique known as the conditional Generative Adversarial Network. Experiments with these techniques yielded two musical pieces—an electronic piano piece and a string quartet—both of which are included in the chapter and deemed to meet professional compositional standards. The chapter concludes by discussing the aesthetic implications of this study and the broader technical and disciplinary challenges surrounding algorithmic music generation.The chapter "Folksong Enthusiasm in Post-Cultural Revolution China" delves into the document Guangxi Folksong Selection (1980), an unpublished record of a field recording trip to Guangxi Province in 1979 by the Class of 1978 Chinese composers. These composers are recognized for their fusion of Western and Chinese musical styles; our understanding of the Chinese elements in their music, however, has primarily been based on the composers’ personal accounts. This chapter offers a more nuanced historical perspective on the composers’ musical “Chineseness” by examining the folksong enthusiasm in post-Cultural Revolution China through a cultural contextualization of the Guangxi Folksong Selection.