Work

Generate Polyphonic Music with Multivariate Masked Autoregressive Flow

Year: 2025

Type: conference-paper

Abstract: This paper explores the use of the Masked Autoregressive Flow (MAF) model for music generation, specifically addressing its limitation to univariate time series.To extend MAF for polyphonic melodies, three approaches are proposed and tested on the Lakh Pianoroll Dataset.The results show promising accuracy and the model's ability to generate original, pleasing melodies, demonstrating the potential of this innovative interdisciplinary approach.

Source: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)

DOI: 10.14428/esann/2025.es2025-86

Authors: Massimiliano Sirgiovanni, Daniele Castellana

Institution: University of Florence

Language: English

Topic: Music Technology and Sound Studies

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: bronze

OpenAlex ID: W4409480570

FWCI:

Cites: 9

Cited by:

Related to: 10