Improved iterative random walk for four-part harmonization
Improved iterative random walk for four-part harmonization
Music generated by random walk from a statistical model generally fails to capture the style and quality of music in its training corpus. In the context of melody harmonization, this paper first demonstrates that violations of general rules of harmony are clearly related to cross-entropy. The paper then describes and evaluates an improved random walk method which efficiently samples low cross-entropy harmonizations. Applying the method, the relationship between cross-entropy and harmonization quality becomes even more apparent. These results will impact on future work in music generation from statistical models.
Whorley, Raymond
801f6e56-9fac-4e56-8fe5-03dd696e6f75
Conklin, Darrell
b53d08f8-3d82-43a5-9523-5b00a4f8f53b
22 June 2015
Whorley, Raymond
801f6e56-9fac-4e56-8fe5-03dd696e6f75
Conklin, Darrell
b53d08f8-3d82-43a5-9523-5b00a4f8f53b
Whorley, Raymond and Conklin, Darrell
(2015)
Improved iterative random walk for four-part harmonization.
Collins, Tom, Meredith, David and Volk, Anja
(eds.)
In Mathematics and Computation in Music: 5th International Conference, MCM 2015 London, UK, June 22-25, 2015 Proceedings.
vol. 9110,
Springer Cham..
(doi:10.1007/978-3-319-20603-5_6).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Music generated by random walk from a statistical model generally fails to capture the style and quality of music in its training corpus. In the context of melody harmonization, this paper first demonstrates that violations of general rules of harmony are clearly related to cross-entropy. The paper then describes and evaluates an improved random walk method which efficiently samples low cross-entropy harmonizations. Applying the method, the relationship between cross-entropy and harmonization quality becomes even more apparent. These results will impact on future work in music generation from statistical models.
This record has no associated files available for download.
More information
Published date: 22 June 2015
Identifiers
Local EPrints ID: 480431
URI: http://eprints.soton.ac.uk/id/eprint/480431
PURE UUID: 08f5c037-962d-470b-b8cb-6ffb07329d90
Catalogue record
Date deposited: 02 Aug 2023 16:38
Last modified: 17 Mar 2024 04:19
Export record
Altmetrics
Contributors
Author:
Raymond Whorley
Author:
Darrell Conklin
Editor:
Tom Collins
Editor:
David Meredith
Editor:
Anja Volk
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics