JAIST Repository >
School of Information Science >
Conference Papers >
Conference Papers >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10119/10933

Title: Structural Similarity Based on Time-span Tree
Authors: Tojo, Satoshi
Hirata, Keiji
Keywords: music analysis
similarity
generative theory of tonal music
time-span tree
Issue Date: 2012-06-22
Publisher: Computer Music Modeling and Retrieval
Magazine name: 9th International Symposium on Computer Music Modeling and Retrieval (CMMR) 2012
Start page: 645
End page: 660
Abstract: Time-span tree is a stable and consistent representation of musicalstructure since most experienced listeners deliver the same one, almost independentlyfrom context and subjectivity. In this paper, we pay attention to the reductionhypothesis of the tree structure, and introduce the notion of distance asa promising candidate of stable and consistent metric of similarity. First, we designa feature structure to represent a time-span tree. Next, we regard that when abranch is removed from the tree the information corresponding to its time-span islost, and suggest that the sum of the length of those removed spans is the distancebetween two trees. We will show that the distance preserves uniqueness in multipleshortest paths, as well as the triangle inequality. Thereafter, we illustrate howthe distance works as a metric of similarity, and then, we discuss the feasibilityand the problem of our methodology.
Rights: All rights remain with the authors. Satoshi Tojo and Keiji Hirata, 9th International Symposium on Computer Music Modeling and Retrieval (CMMR) 2012, pp.645-660.
URI: http://hdl.handle.net/10119/10933
Material Type: author
Appears in Collections:b11-1. 会議発表論文・発表資料 (Conference Papers)

Files in This Item:

File Description SizeFormat
18142.pdf1019KbAdobe PDFView/Open

All items in DSpace are protected by copyright, with all rights reserved.

 


Contact : Library Information Section, Japan Advanced Institute of Science and Technology