We present a method to classify audio recordings of folk songs into tune families. For this, we segment both the query recording and the recordings in the collection. The segments can be used to relate recordings to each other by evaluating the recurrence of similar melodic patterns. We compare a segmentation that results in what can be considered cognitive units to a segmentation into segments of fixed length. It appears that the use of ‘cognitive’ segments results in higher classification accuracy. 1. BACKGROUND Large collections of monophonic folk song recordings are interesting from a music cognition perspective since they represent musical performances of common people. Most people share a ‘common core of musical knowledge’ (Peretz 2006: section 2). Since recorded folk songs were sung from memory, knowledge about the process of remembering and reproducing melodies can be used to employ these recordings in the context of folk song research, music information retrieval or music cognition studies. This study combines ideas and approaches from etnomusicology, music cognition and computer science. One of the research questions of etnomusicology is how melodies in an oral tradition relate to each other (Nettl 2005, chapter 9; Van Kranenburg et al. 2009a). Samuel Bayard (1950) developed the concept tune family to denote a group of melodies that share a common origin, which, in the most simple case, is a single tune. The idea that melodies from the same tune family are related by shared melodic motifs has a long history in folk song research. Nettl (2005: p. 117f) discusses the relative independence of shorter units of musical thought. These might ‘wander’ from melody to melody and from country to country (Tappert 1890). Marcello Keller (1988) explains the relations between Trentino folk music compositions by means of a repertoire of ‘segments’ that is used in the act of composing. To cope with specific relations between melodies he encountered in Irish folk music, James Cowdery (1984) extended Bronson’s concept of tune family by including melodies that are related by sharing melodic material from the same ‘pool of motives’. Finally, one of the conclusions from a previous study on the same corpus of songs that we use in this paper is that recurring motifs are more important than contour and rhythm for recognizing a song (Volk et al. 2008).
[1]
Wilhelm Tappert,et al.
Wandernde Melodien : eine musikalische Studie
,
1965
.
[2]
Hideki Kawahara,et al.
YIN, a fundamental frequency estimator for speech and music.
,
2002,
The Journal of the Acoustical Society of America.
[3]
Emilios Cambouropoulos,et al.
Towards a General Computational Theory of Musical Structure
,
1998
.
[4]
R. Jackendoff,et al.
A Generative Theory of Tonal Music
,
1985
.
[5]
B. Bronson,et al.
Prolegomena to a Study of the Principal Melodic Families of British-American Folk Song
,
1950
.
[6]
Remco C. Veltkamp,et al.
A Manual Annotation Method for Melodic Similarity and the Study of Melody Feature Sets
,
2008,
ISMIR.
[7]
M S Waterman,et al.
Identification of common molecular subsequences.
,
1981,
Journal of molecular biology.
[8]
E. Narmour.
The analysis and cognition of basic melodic structures
,
1992
.
[9]
I. Peretz.
The nature of music from a biological perspective
,
2006,
Cognition.
[10]
O. Gotoh.
An improved algorithm for matching biological sequences.
,
1982,
Journal of molecular biology.
[11]
Bruno Nettl.
The Study Of Ethnomusicology
,
1983
.
[12]
Nicholas Cook,et al.
Empirical musicology : aims, methods, prospects
,
2004
.
[13]
Remco C. Veltkamp,et al.
MUSICAL MODELS FOR FOLK-SONG MELODY ALIGNMENT
,
2009
.
[14]
G. A. Miller.
THE PSYCHOLOGICAL REVIEW THE MAGICAL NUMBER SEVEN, PLUS OR MINUS TWO: SOME LIMITS ON OUR CAPACITY FOR PROCESSING INFORMATION 1
,
1956
.
[15]
Frans Wiering,et al.
Modelling Folksong Melodies
,
2009
.
[16]
Remco C. Veltkamp,et al.
Musical Models for Melody Alignment
,
2009,
ISMIR.
[17]
J. Cowdery.
A Fresh Look at the Concept of Tune Family
,
1984
.
[18]
Wilhelm Tappert.
Wandernde Melodien. Eine Musikalische Studie von Wilhelm Tappert
,
1890
.
[19]
R. Veltkamp,et al.
Collaboration Perspectives for Folk Song Research and Music Information Retrieval: The Indispensable Role of Computational Musicology
,
2009
.