The Open UniversitySkip to content

An AI Tool for the Analysis and Generation of Melodies

Smith, Matthew and Holland, Simon (1992). An AI Tool for the Analysis and Generation of Melodies. In: 1992 International Computer Music Conference (ICMC 1992), 14-18 Oct 1992, San Jose State University, California, International Computer Music Association, pp. 61–64.

Full text available as:
PDF (Version of Record) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (234kB) | Preview
Google Scholar: Look up in Google Scholar


Artificial Intelligence (AI) offers music theorists and cognitive musicologists the means to express theories unambiguously, through the implementation of computational models. Any inconsistencies in a theory will be discovered during the encoding process, and, since the computer has no musical intuitions or experience, any obvious or assumed knowledge must be made explicit by the knowledge engineer.

Our eventual goal is the development of an intelligent computer environment for students learning melody composition. This paper describes the development of one tool which will be used within the tutoring system: a program for the analysis and generation of melodies, based upon Eugene Narmour's (1990) ImplicationRealisation Model - a hypothetical theory for melody analysis.

The tool is being developed using the declarative, logic-based language Prolog. In addition to encoding Narmour's analytical theory, we have designed extensions to the computational model which allow the generation of melodies using constraint-based techniques. This means that with a single computer model a user will be able to analyse a given melody, modify a melody in musically 'sensible' ways (explained below) by changing notes at various hierarchical levels, generate new melodies and finish partially completed melodies.

The paper is presented in four sections: the first describes those aspects of Narmour's theory necessary to understand the computational model; the second section describes the design of the analytical model; the third outlines how the computational model can be used to test Narmour's theory; a final section describes the additions to the model to allow generation of melodies using constraint techniques.

Item Type: Conference or Workshop Item
Copyright Holders: 1992 The Authors
ISSN: 2223-3881
Keywords: artificial intelligence; music analysis; music composition; melody; symbolic programming
Academic Unit/School: Faculty of Science, Technology, Engineering and Mathematics (STEM) > Computing and Communications
Faculty of Science, Technology, Engineering and Mathematics (STEM)
Related URLs:
Item ID: 59695
Depositing User: Simon Holland
Date Deposited: 01 Apr 2019 08:49
Last Modified: 16 Apr 2019 13:35
Share this page:

Download history for this item

These details should be considered as only a guide to the number of downloads performed manually. Algorithmic methods have been applied in an attempt to remove automated downloads from the displayed statistics but no guarantee can be made as to the accuracy of the figures.

Actions (login may be required)

Policies | Disclaimer

© The Open University   contact the OU