The University of Edinburgh -
Division of Informatics
Forrest Hill & 80 South Bridge

MSc Thesis #9748

Title:Statistical Music Modelling
Date:Sep 1997
Abstract:This project models music using statistical techniques. Ideas are borrowed from statistical natural language processing. Trigram probabilities are estimated on a corpus of seventeenth-century dance music. The probabilities are then used for experiments in evaluation (assigning probabilities to pieces) and generation (composition). The corpus is preprocessed in a novel way by converting the harmonies into a normal form to capture the underlying harmonic similarities between pieces. It is then automatically marked up with constituent boundaries (beginnings and ends of pieces, phrases and bars), to enable the trigram learning process to capture some of the structure of the music.

[Search These Pages] [DAI Home Page] [Comment]