Please use this identifier to cite or link to this item:
Title: Content-based music retrieval by acoustic query
Keywords: music information retrieval content based humming
Issue Date: 16-Sep-2004
Citation: ZHU YONG WEI (2004-09-16). Content-based music retrieval by acoustic query. ScholarBank@NUS Repository.
Abstract: This thesis presents a novel methodology for content-based music retrieval using acoustic (humming) melody queries. The problems to be solved are representation, extraction and matching of melodies from either symbolic music input or acoustic music input, considering inevitable errors or variations in the humming query signals. The proposed techniques include a robust melody representation, called pitch line, a robust pitch extraction method for construction of pitch line from humming input, melody similarity metrics based on pitch line, and three melody matching techniques involving key transposition and melody alignment. Novel melody features and matching methods using melody slope and melody skeleton have been proposed for robust and efficient key transposition and melody alignment. A music scale modelling method has been proposed for very efficient key transposition for melody matching. Experimental results have shown the effectiveness of the proposed methods for music retrieval using acoustic queries.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
HD981992E-PHD-THESIS.pdf2.67 MBAdobe PDF



Page view(s)

checked on Jul 28, 2018


checked on Jul 28, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.