Home > Research > Publications & Outputs > Automatic Derivation of Musical Structure: A To...
View graph of relations

Automatic Derivation of Musical Structure: A Tool for Research on Schenkerian Analysis.

Research output: Contribution to conference - Without ISBN/ISSN Other

Publication date23/09/2007
Number of pages4
<mark>Original language</mark>English
EventInternational Conference on Music Information Retrieval - Vienna
Duration: 1/09/2007 → …


ConferenceInternational Conference on Music Information Retrieval
Period1/09/07 → …

Bibliographic note

Since the advent of musical computing in the 1970s, several researchers have attempted to implement Schenkerian analysis in a computer program, or the related style of analysis of Lerdahl & Jackendoff. All have faced serious difficulties, in particular (1) 'combinatorial explosion' where the number of possibilities to be considered becomes intractably large, and (2) poorly defined criteria for selecting a 'good' analysis from among the possibilities. For the first time, this paper presented a computer program to derive the structure of a piece of music from its score in a manner which overcomes the first problem without relying either on an expert user to adjust parameters or on the application of probabilities. It is also the first to apply to a full polyphonic score. The program takes unstructured symbolic input such as a MIDI file and produces a matrix of possible reductions according to the system described in Marsden's item 2 as its output. Demonstration software may be seen online along with an extended version of the paper giving more detail than possible in the space allowed in the conference proceedings. The result is a tool for research on the second problem, to discover the criteria for selecting good analyses, and a promise of music-analysis software and automatic music-information-retrieval systems which look below the surface of a piece to deal with the underlying musical patterns and structure. Other outcomes: papers presented at the Dublin International Music Analysis Conference (2005) and the International Computer Music Conference (2005) presented an early stage of this work to other audiences. RAE_import_type : Conference contribution RAE_uoa_type : LICA