Modelling a listener's perception of musical rhythm requires both an understanding of rhythm as a whole as well as a definition of its constituent elements. The hypothesis is that once we can adequately define rhythm, we can then begin to design and implement models to gain insight into the perceptual processes which occur when listening to rhythmic sequences. This research outlines studies which have attempted to define and outline both the structure and the perception of rhythm. Based on the conclusions of these investigations, a computer model is designed and implemented using connectionist techniques. The emphases on this model are to arrive at a viable solution for extracting rhythmic material from performed input, and to implement time-scale invariance. Time-scale invariance allows the system to recognize (categorize) similar patterns played at different tempos as being the same pattern. The performance of this model is evaluated against earlier models designed with similar neural network architectures as well as in relation to the conclusions drawn by music theorists and psychologists.
Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.55412 |
Date | January 1993 |
Creators | Hogan, Kharim Manuelle |
Contributors | Pennycook, Bruce (advisor) |
Publisher | McGill University |
Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
Language | English |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Format | application/pdf |
Coverage | Master of Arts (Faculty of Music.) |
Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
Relation | alephsysno: 001458491, proquestno: AAIMM05516, Theses scanned by UMI/ProQuest. |
Page generated in 0.0018 seconds