Exploring Tokenization Methods for Multitrack Sheet Music Generation
Sound
2024-10-24 v1 Artificial Intelligence
Audio and Speech Processing
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.
Keywords
Cite
@article{arxiv.2410.17584,
title = {Exploring Tokenization Methods for Multitrack Sheet Music Generation},
author = {Yashan Wang and Shangda Wu and Xingjian Du and Maosong Sun},
journal= {arXiv preprint arXiv:2410.17584},
year = {2024}
}
Comments
3 pages, 1 figure, 1 table