Exploring Tokenization Methods for Multitrack Sheet Music Generation
Yashan Wang (Central Conservatory of Music)*, Shangda Wu (Central Conservatory of Music), Xingjian Du (University of Rochester), Maosong Sun (Tsinghua University)
This paper will be presented in person
Abstract:
This study explores the tokenization of multitrack sheet music in ABC notation, introducing two methods—bar-stream and line-stream patching. We compare these methods against existing techniques, including bar patching, byte patching, and Byte Pair Encoding (BPE). In terms of both computational efficiency and the musicality of the generated compositions, experimental results show that bar-stream patching performs best overall compared to the others, which makes it a promising tokenization strategy for sheet music generation.