Generative Modeling of Bach-Style Symbolic Music: A Comparative Study of Autoregressive, Latent-Variable, and Adversarial Approaches

Kyuil Lee, Dezhi Yu, Yongkang Huang|June 11, 2026arXiv

Key Takeaway

For symbolic music generation, autoregressive models with attention outperform VAEs and GANs at producing stylistically coherent Bach compositions, though vector quantization significantly improves VAE performance by preventing posterior collapse.

Summary

This paper compares three types of AI models for generating Bach-style piano music from symbolic notation: autoregressive models (which predict one note at a time), latent-variable models (which learn compressed representations), and adversarial models (which compete to fool each other).

architecture evaluation applications

Key Terms

autoregressive-model variational-autoencoder vector-quantization posterior-collapse generative-adversarial-network