A Joyful Ode to Automatic Orchestration

By François Pachet.

ACM Transactions on Intelligent Systems and Technology 8(2), Article 18, 13 pages, October 2016. DOI: 10.1145/2897738.

Read the paper (PDF)
Publisher record

About the paper

This paper reports an experiment in reorchestrating Beethoven’s Ode to Joy in seven styles using techniques developed in the Flow Machines project. A concrete musical commission brings multiple generation and production tasks together and exposes the challenges of producing complete arrangements.

Abstract summary

The article examines what happens when music-generation methods must work together toward a finished musical result. The shared goal connects style modelling, orchestration and production, and reveals practical research questions that emerge during the creation of the arrangements.

Musical demonstration

Watch Machine Learning Techniques for Reorchestrating the European Anthem — a demonstration of the arrangements, credited to François Pachet, Pierre Roy, Mathieu Ramona, Marco Marchini, Gaëtan Hadjeres, Emmanuel Deruty, Benoît Carré and Fiammetta Ghedini. The video received the AAAI 2016 Best Video Award.

Related work

Flow Machines project
Assisted Music Creation with Flow Machines: Towards New Categories of New
Maximum Entropy Models Capture Melodic Styles

Back to Publications