MusPy: A Toolkit for Symbolic Music Generation
This toolkit addresses the need for standardized tools in symbolic music generation research, but it is incremental as it builds on existing concepts without introducing new methods.
The authors introduced MusPy, an open-source Python toolkit for symbolic music generation, providing tools for dataset management, preprocessing, and evaluation, and conducted cross-dataset experiments showing domain overlap and dataset representativeness.
In this paper, we present MusPy, an open source Python library for symbolic music generation. MusPy provides easy-to-use tools for essential components in a music generation system, including dataset management, data I/O, data preprocessing and model evaluation. In order to showcase its potential, we present statistical analysis of the eleven datasets currently supported by MusPy. Moreover, we conduct a cross-dataset generalizability experiment by training an autoregressive model on each dataset and measuring held-out likelihood on the others---a process which is made easier by MusPy's dataset management system. The results provide a map of domain overlap between various commonly used datasets and show that some datasets contain more representative cross-genre samples than others. Along with the dataset analysis, these results might serve as a guide for choosing datasets in future research. Source code and documentation are available at https://github.com/salu133445/muspy .