Tune retrieval based on the similarity of melody

Authors:
Yuka Nakagawa;Yasuaki Nakamura
Affiliations:
Department of Computer Science, Hiroshima City University;Department of Computer Science, Hiroshima City University
Venue:
WEDELMUSIC'02 Proceedings of the Second international conference on Web delivering of music
Year:
2002

Citing 3
Cited 0

Query by humming: musical information retrieval in an audio database

Proceedings of the third ACM international conference on Multimedia
Tune Retrieval in the Multimedia Library

Multimedia Tools and Applications
Similarity Indexing with the SS-tree

ICDE '96 Proceedings of the Twelfth International Conference on Data Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

A melody-based retrieval of music, which is one type of the content-based retrieval, is proposed in the paper. Every fragment of a melody is represented as an appropriate N-dimensional vector, called a feature vector. A feature vector consists of values corresponding to intervals and rhythm of a melody. Depending on the distance between every fragment of a melody in the music database to the entered melody, the k-nearest melodies can be obtained quickly by using the SS-tree. A prototype system is developed to demonstrate the effectiveness of our approach.