Options
Symbolic representation of chords for rule-based evaluation of tonal progressions
Citation
Chong, E. K. M., & Ding, Q. (2014). Symbolic representation of chords for rule-based evaluation of tonal progressions. eContact!, 16(2). http://cec.sonus.ca/econtact/16_2/chong-ding_representation.html
Abstract
Whilst a number of approaches to representing chords in a Classical tonal context have been proposed for various computational or data retrieval purposes, the present project differs in its being motivated and shaped by musicpedagogical intentions, which include offering automated error feedback. For the same reason, for the evaluation of chord progressions, a rule-based method using declarative rules was adopted instead of a data-driven, pattern-discovery one. Both the encoding system and the formulated rules are based on Classical theory of functional harmonic voice leading. The ultimate aim is to develop a musically intelligent interactive system that automatically evaluates tonal progressions and provides assessment feedback for the purpose of teaching Western Classical tonal theory. In a nutshell, the computational system’s input are chords drawn from the major-minor tonal system. Roman numerals with figured-bass indications (e.g., I6, viio4/2) are the chord symbols to be encoded. The encoding reflects a number of pertinent theoretical elements: 1) the key context; 2) the scale degree of the chord; 3) the chord type; and 4) the chord inversion. A rule engine, based on JBoss Drools and the Rete algorithm, is then designed to evaluate chord progressions based on considerations of root motion, bass movement and other tonal voice-leading factors. The musical intelligence of this system therefore simulates human musical thinking as encapsulated in a typical undergraduate theory of harmony.
Date Issued
2014
Publisher
Canadian Electroacoustic Community
Journal
eContact!