Music Score Expansion with Variable-Length Infilling
This addresses a domain-specific problem for music generation, but it is incremental as it applies an existing method to a new task.
The paper tackled the problem of expanding musical segments by using a variable-length infilling model to prolong 20 segments from 12 to 16 bars, and found that the model preserves musical boundaries effectively based on objective metrics like Register Histogram Similarity.
In this paper, we investigate using the variable-length infilling (VLI) model, which is originally proposed to infill missing segments, to "prolong" existing musical segments at musical boundaries. Specifically, as a case study, we expand 20 musical segments from 12 bars to 16 bars, and examine the degree to which the VLI model preserves musical boundaries in the expanded results using a few objective metrics, including the Register Histogram Similarity we newly propose. The results show that the VLI model has the potential to address the expansion task.