Publications

Featured Publication

Machine learning of artistic fingerprints in jazz

Cheston, H., Bance, R., & Harrison, P. M. C. (2026). Machine learning of artistic fingerprints in jazz. Nature Machine Intelligence, 8(8), 1261–1274.

Artists are often recognizable through distinctive patterns—“fingerprints”—in their work. This paper trains machine-learning models to identify 20 iconic jazz pianists from 84 hours of recordings, including a multi-input architecture that analyses melody, harmony, rhythm, and dynamics separately. The best model reaches 94% accuracy across 20 classes, and the models reveal which musical elements most strongly distinguish individual artists.

Open-source model code and an interactive web app accompany the paper.

Read the paper →
Machine learning of artistic fingerprints in jazz

2026

2025

2024

2023

2022

2021

2020

2019

2018

2016

2015