Multiculturality

From MIReS

Latest revision as of 18:16, 20 April 2013

Most music makes very little sense unless we experience it in its proper cultural context, thus the processing of music information has to take into account this cultural context. Most of MIR has focused on the mainstream popular Western music of the past few decades and thus most research results and technologies have a cultural bias towards that particular cultural context. The challenge is to open up our view on music, to develop technologies that take into account the existing musical diversity and thus the diverse musical cultural contexts. To approach the multicultural aspects of MIR there is a need to involve researchers from both engineering disciplines (Signal Processing, Machine Learning) and humanities (Musicology, Cultural Studies), and to involve people belonging to the specific cultures being studied. This approach will offer the possibility to identify new MIR problems and methodologies that could impact the whole MIR field. At the same time the development of Information Technologies that reflect diversity should help preserve the cultural richness of our world, which is threatened by the globalisation and homogenisation of the IT infrastructures. This is a topic that has started to be addressed by the MIR community but that will require much bigger efforts, not just by the research community but by political and industrial bodies.

Back to → Roadmap:Socio-cultural perspective

State of the art

MIR uses a variety of methodologies, but the most common approximations are based on using signal processing and machine learning methods that treat musical data as any other machine readable data, thus without much domain knowledge. On the other hand the research done within the fields of Computational Musicology and Computational Ethnomusicology puts a special emphasis on the musical and cultural aspects, thus incorporating domain knowledge that we want to emphasise here. These two research areas have been growing in the last few years and their influence in the MIR community has been increasing (see section Knowledge-driven methodologies: Musicology).

The term Computational Musicology comes from the research tradition of musicology, a field that has focused on the study of the symbolic music representations (scores) of the classical western music tradition [Camilleri, 1993]. This research perspective takes advantage of the availability of scores in machine-readable format and of all the musicological research that has been done on this music tradition. Music theoretical models are very much followed and current research focuses on the understanding and modelling of different musical facets such as melody, harmony or structure of western classical music. This research can be followed in the yearly journal Computing in Musicology [Hewlett and Selfridge-Field, 1991]. From these references it can be observed that this field has been opening up, approaching other types of music, such as popular western music or different world music traditions, and it has started to use other types of data sources, such as audio recording.

In [Tzanetakis et al., 2007] the concept of Computational Ethnomusicology was introduced to refer to the use of computer tools to assist in ethnomusicological research. The emphasis is on the study of folk and popular music traditions that are outside the western classical and pop cultures, thus cultures that tend to be based on oral traditions and that have been mainly studied through audio recordings. Since the article was published there has been an increasing number of research articles related to Computational Ethnomusicology. For instance, according to [Cornelis et al., 2010], the percentage of papers on this area at the annual ISMIR conference increased from 4.8% in 2002 to 8.1% in 2008. A year later, in 2009, ISMIR hosted an oral session devoted to the analysis of folk music, sociology and ethnomusicology. After this event, a group of researchers working on MIR and ethnomusicology started the EthnoMIR discussion group which has organised a yearly workshop on Folk Music Analysis (2011 in Athens, 2012 in Seville, and 2013 in Amsterdam) with the purpose of gathering researchers who work in the area of computational music analysis of music from different cultures, using symbolic or signal processing methods, to present their work, discuss and exchange views on the topic. At the ISMIR 2011 there was a session dedicated to "non-western music" and at ISMIR 2012 there was a session on "musical cultures" and a larger than ever amount of contributions related to different musical traditions. In 2011 the European Research Council funded a project entitled "CompMusic: Computational Models for the discovery of the world's music" that is studying five art music traditions (Hindustani, Carnatic, Turkish-makam, Andalusi, and Beijing Opera) from an MIR perspective.

Recent studies on non-western music show the need to expand or even rethink some of the MIR methodologies. Some papers deal with specific musical facets, such as timbre/instrumentation [e.g. Proutskova and Casey, 2009], rhythm [e.g. Holzapfel and Stylianou, 2011], motives [e.g. Lartillot and Ayari, 2006; Conklin and Anagnostopoulou, 2010], tuning and scale [e.g. Gedik and Bozkurt, 2009; Moelants et al., 2009], melody [e.g. Wiering et al., 2009; Mora et al., 2010], or performance variations [e.g. Müller et al., 2010; Henbing and Leman, 2007]. From these references it becomes clear that many of the musical concepts used in MIR need to be rethought and new approaches developed if we want to take a multicultural perspective. Concepts like tuning, rhythm, melody, scale, chord, tonic, … are very culture specific, and need to be treated as such. Among the non-western music repertoires that have been most studied from this perspective are the Turkish-makam [e.g. Bozkurt, 2008] and the art traditions of India [e.g. Krishnaswamy, 2004; Koduri et al., 2012].

References

• [Bozkurt, 2008] Bariş Bozkurt. An Automatic Pitch Analysis Method for Turkish Maqam Music. Journal of New Music Research, 37(1):1-13, March 2008.

• [Camilleri, 1993] Lelio Camilleri. Computational Musicology A Survey on Methodologies and Applications. Revue Informatique et Statistique dans les Sciences humaines, pp. 51-65, 1993.

• [Conklin and Anagnostopoulou, 2010] Darrell Conklin and Christina Anagnostopoulou. Comparative Pattern Analysis of Cretan Folk Songs. Journal of New Music Research, 40(2): 119-125, 2010.

• [Cornelis et al., 2010] Olmo Cornelis, Micheline Lesaffre, Dirk Moelants, and Marc Leman. Access to Ethnic Music: Advances and Perspectives in Content-Based Music Information Retrieval. Signal Processing, 90(4): 1008-1031, 2010.

• [Gedik and Bozkurt, 2009] Ali Gedik and Bariş Bozkurt. Evaluation of the Makam Scale Theory of Arel for Music Information Retrieval on Traditional Turkish Art Music. Journal of New Music Research, 38(2):103-116, 2009.

• [Henbing and Leman, 2007] L. Henbing and Marc Leman. A Gesture-Based Typology of Sliding-Tones in Guqin Music. Journal of New Music Research, pp. 61-82, 2007.

• [Holzapfel and Stylianou, 2011] Andre Holzapfel and Yannis Stylianou. Scale Transform in Rhythmic Similarity of Music. IEEE Transactions on Audio, Speech & Language Processing, 19(1): 176-185, 2011.

• [Koduri et al., 2012] Gopala Krishna Koduri, Sankalp Gulati, Preeti Rao, and Xavier Serra. Rāga Recognition based on Pitch Distribution Methods. Journal of New Music Research, 41(4): 337-350, 2012.

• [Krishnaswamy, 2004] Arvindh Krishnaswamy. Melodic Atoms for Transcribing Carnatic Music. In Proceedings of the 5th International Conference on Music Information Retrieval, pp. 1-4, Barcelona, Spain, 2004.

• [Lartillot and Ayari, 2006] Olivier Lartillot and Mondher Ayari. Motivic Pattern Extraction in Music, and Application to the Study of Tunisian Modal Music. South African Computer Journal, 36: 16-28, 2006.

• [Moelants et al., 2009] Dirk Moelants, Olmo Cornelis, and Marc Leman. Exploring African tone scales. In Proceedings of the 10th International Society for Music Information Retrieval Conference, pp. 489-494, Kobe, Japan, 2009.

• [Mora et al., 2010] Joaquín Mora, Francisco Gómez, Emilia Gómez, Francisco Escobar-Borrego, and José Miguel Díaz-Báñez. Characterization and Melodic Similarity of A Cappella Flamenco Cantes. In Proceedings of the 11th International Society for Music Information Retrieval Conference, Utrecht, Netherlands, 2010.

• [Müller et al., 2010] Meinard Müller, Peter Grosche, and Frans Wiering. Automated analysis of performance variations in folk song recordings. In Proceedings of the International Conference on Multimedia Information Retrieval, New York, USA, 2010. ACM Press.

• [Proutskova and Casey, 2009] Polina Proutskova and Michael Casey. You Call That Singing? Ensemble Classification for Multi-Cultural Collections of Music Recordings. In Proceedings of the 10th International Society for Music Information Retrieval Conference, pp. 759-764, Kobe, Japan, 2009.

• [Tzanetakis et al., 2007] G. Tzanetakis, A. Kapur, W.A. Schloss, and M. Wright. Computational Ethnomusicology. Journal of Interdisciplinary Music Studies, 1(2): 1-24, 2007.

• [Hewlett and Selfridge-Field, 1991] Walter B. Hewlett and Eleanor Selfridge-Field. Computing in Musicology, 1966-91. Computers and the Humanities, 25(6): 381-392, 1991.

• [Wiering et al., 2009] Frans Wiering, Remco C Veltkamp, Jörg Garbers, Anja Volk, Peter van Kranenburg, and Louis P Grijp. Modelling Folksong Melodies. Interdisciplinary Science Reviews, 34(2): 154-171, 2009.

Challenges

• Identify and characterise music cultures that can be studied from a data driven perspective. For MIR purposes a musical culture can be considered a combination of a user community plus a musical repertoire that can be characterised computationally. Thus we can extract data from the trace left online by a user community, such as online social networks, and from different music data repositories created by that community, especially audio and scores. With this type of data we can then study quite a few aspects of a given musical culture. The challenge is to identify musical cultures that can be studied like this.

• Gather and make available culturally relevant data for different music cultures. Gather different data sources (audio, audio descriptors, editorial metadata, expert data, user commentaries, ...) with which to study and characterise the community+repertoire of the selected cultures. This data has to be made available to the research community.

• Identify specific music characteristics for each culture. Identify particular semantic music concepts and characteristics that are specific to each culture. These should be the aspects that allow us to differentiate the different musical cultures.

• Develop methodologies for culture specific problems. Develop knowledge based data processing approaches that can take advantage of the specificities of each culture, thus modeling the characteristics of each user community and music repertoire.

• Develop specific applications of relevance for each cultural context. The members of each user community might have specific needs and thus the applications to be developed for them should target their context and interests.

• Carry out comparative studies using computational approaches. These comparative studies should be done from the research results obtained in the characterisation and modeling of specific music traditions and repertoires.

Back to → Roadmap:Socio-cultural perspective