2010 National Conference of Higher Vocational and Technical Education on Computer Information (NCHVT 2010 E-BOOK)

Rizhao,China,12.4-12.5,2010

ISBN: 978-1-935068-31-0 Scientific Research Publishing, USA

E-Book 358pp Pub. Date: December 2010

Category: Social Sciences & Humanities

Price: $70

Title: Algorithms of Internet Music Search Engine by Humming and Performance Evaluation
Source: 2010 National Conference of Higher Vocational and Technical Education on Computer Information (NCHVT 2010 E-BOOK) (pp 148-153)
Author(s): Yun-feng Dong, Computing Center, Shandong Institute of Light Industry, 250353,Jinan,China
Bei Qi, Network Center, Shandong Institute of Light Industry, 250353,Jinan,China
Abstract: This paper have made a study of algorithms that commonly used in internet search engine by humming, and evaluated the performance of algorithms in the large-scale music search system by humming. In order to support study of various approximate match algorithms’ compare, an algorithm-independent test platform based on network was developed to evaluate the performance of algorithms. The basic algorithms include Suffix-tree, Hidden Markov Models (HHM), approximation melody algorithm, Dynamic Time Wrapping (DTW) and similarity matching algorithm. The author have carried out large amount of experiments on the algorithms of approximation melody algorithm, Dynamic Time Wrapping (DTW) and similarity matching algorithm, analyzed and evaluated them. The experiment results indicate that, similarity matching algorithm is fit to internet music search engine by humming.
Free SCIRP Newsletters
Copyright © 2006-2024 Scientific Research Publishing Inc. All Rights Reserved.
Top