Identifying Prototypical Melodies by Extracting Approximate Repeating Pattern from Music Works

The concept of a prototypical melody has been proposed to characterize sets of similar musical segments in a composition. In musicology, the degree of importance associated with a prototypical melody is proportional to the number of musical segments similar to it. In this paper, a novel approach is developed to extract all the prototypical melodies in a musical work. Our approach considers each music segment as a prototypical melody candidate and utilizes edit distance to isolate a set of music segments that are similar to this candidate. To expedite the process, a lower bounding mechanism is used to estimate the number of similar musical segments for each candidate to eliminate impossible options. Furthermore, the approach is extended to facilitate the extraction of all prototypical melodies in a set of musical works. Analysis is carried out on a real data set, and the results indicate significantly improved performance in average response time relative to existing approaches.

[1]  E. Narmour.  The analysis and cognition of basic melodic structures , 1992 .

[2]  Arbee L. P. Chen,et al.  Music Classification Using Significant Repeating Patterns , 2004, DASFAA.

[3]  Eugene L. Lawler,et al.  Sublinear approximate string matching and biological applications , 1994, Algorithmica.

[4]  Anna Pienimäki,et al.  Indexing Music Databases Using Automatic Extraction of Frequent Phrases , 2002, ISMIR.

[5]  Manolis Christodoulakis,et al.  A Fast, Randomised, Maximal Subset Matching Algorithm for Document-Level Music Retrieval , 2006, ISMIR.

[6]  Gonzalo Navarro,et al.  Indexed Hierarchical Approximate String Matching , 2008, SPIRE.

[7]  Maxime Crochemore,et al.  Algorithms on strings , 2007 .

[8]  Shrikanth Narayanan,et al.  Automatic main melody extraction from midi files with a modified Lempel-Ziv algorithm , 2001, Proceedings of 2001 International Symposium on Intelligent Multimedia, Video and Speech Processing. ISIMP 2001 (IEEE Cat. No.01EX489).

[9]  Arbee L. P. Chen,et al.  Discovering nontrivial repeating patterns in music data , 2001, IEEE Trans. Multim..

[10]  Gonzalo Navarro,et al.  A guided tour to approximate string matching , 2001, CSUR.

[11]  Hans-Peter Kriegel,et al.  The R*-tree: an efficient and robust access method for points and rectangles , 1990, SIGMOD '90.

[12]  Johan Sundberg,et al.  Common Secrets of Musicians and Listeners - An analysis-by-synthesis Study of Musical Performance , 1991 .

[13]  Eleanor Selfridge-Field,et al.  Melodic Similarity : concepts, procedures, and applications , 1998 .

[14]  Steffen Pauws,et al.  CubyHum: a fully operational "query by humming" system , 2002, ISMIR.

[15]  Alan Fern,et al.  Learning for efficient retrieval of structured data with noisy queries , 2007, ICML '07.

[16]  Pierre-Yves Rolland.  FIExPat: flexible extraction of sequential patterns , 2001, Proceedings 2001 IEEE International Conference on Data Mining.

[17]  Maxime Crochemore,et al.  A Pattern Extraction Algorithm for Abstract Melodic Representations that Allow Partial Overlapping of Intervallic Categories , 2005, ISMIR.

[18]  Dan Gusfield,et al.  Algorithms on Strings, Trees, and Sequences - Computer Science and Computational Biology , 1997 .

[19]  Dan Gusfield.  Algorithms on Strings, Trees, and Sequences - Computer Science and Computational Biology , 1997 .

[20]  William Forde Thompson.  Melodic Similarity: Concepts, Procedures, and Applications (Computing in Musicology 11) Walter B. Hewlett Eleanor Selfridge-Field , 1999 .