Final Report Summary - COMPMUSIC (Computational models for the discovery of the world's music)
We studied five music cultures: Hindustani (North India), Carnatic (South India), Turkish-makam (Turkey), Arab-Andalusian (Maghreb), and Beijing Opera (China). The research team included members from each of the cultures being studied.
Within the field of MIR there are many research problems that can benefit from a culture specific perspective. CompMusic focused on the extraction of features from audio music recordings related to melody and rhythm, and on the semantic analysis of the contextual information of those recordings. The goal was to characterize culture specific musical facets of each repertoire and to develop musically meaningful similarity measures with them.
The research in CompMusic is data driven, thus it revolves around corpora. We constructed a research corpus for each music tradition. The types of data gathered were mainly audio recordings and editorial metadata, which were then complemented with descriptive information about the items we have, and in some cases with music scores and/or lyrics.
We focused on the study of melody and rhythm to develop methodologies with which to characterize and explore each music repertoire. Most contributions resulted from developing signal processing approaches to characterize the basic melodic and rhythmic features of each tradition, using audio and scores. We also applied pattern matching and semantic analysis techniques to characterize and discover relationships between the different musical concepts that we identified, using both metadata accompanying the recordings and community generated information.
In order to evaluate our research results we defined various user scenarios and have developed a framework, Dunya, comprising data and tools, with which users can interact and with which we can disseminate most of the research result.
The major dissemination strategy of CompMusic has to be open about everything we do; sharing our ideas, goals, and results as openly and widely as we can. All our publications have been made available as soon as they have been written, all our code is open source, and all the data we generated is available under open licenses. Everything is available through: http://compmusic.upf.edu