LZMidi: Compression-Based Symbolic Music Generation

Abhiram Rao Gorle

Abstract:

Recent advances in symbolic music generation pri-
marily rely on deep learning models such as Transformers,
GANs, and diffusion models. While these approaches achieve
high-quality results, they require substantial computational
resources, limiting their scalability. We introduce LZMidi, a
lightweight symbolic music generation framework based on a
Lempel-Ziv (LZ78)-induced sequential probability assignment
(SPA). By leveraging the discrete and sequential structure of
MIDI data, our approach enables efficient music generation on
standard CPUs with minimal training and inference costs. The-
oretically, we establish universal convergence guarantees for our
approach, underscoring its reliability and robustness. Compared
to state-of-the-art diffusion models, LZMidi achieves competi-
tive Fréchet Audio Distance (FAD), Wasserstein Distance (WD),
and Kullback-Leibler (KL) scores, while significantly reducing
computational overhead—up to 30× faster training and 300×
faster generation. Our results position LZMidi as a significant
advancement in compression-based learning, highlighting how
universal compression techniques can efficiently model and
generate structured sequential data, such as symbolic music,
with practical scalability and theoretical rigor.