Note: This bibliographic page is archived and will no longer be updated. For an up-to-date list of publications from the Music Technology Group see the Publications list .

Digital encodings of early music: usability factors for computational analysis

Title Digital encodings of early music: usability factors for computational analysis
Publication Type Conference Paper
Year of Publication 2015
Conference Name International Workshop on Musicology and Informatics in Dialogue
Authors Martorell, A.
Conference Start Date 17/09/2015
Conference Location Saarbrücken (Germany)
Abstract Computational music analysis materializes in two different formats, namely one that aims to provide definitive results and another that only facilitates information encouraging analytical inquiry. Departing from the premise that analytical conclusions are to be ultimately derived by human analysts, the musical data to be processed should encode relevant features of analytical interest, whereas the output information has to be easily usable. I will draw on two contrasting database initiatives, arguing that, besides the appropriate encoding choices, the modalities of access to the data and the presentation of the output results are of utmost importance. The Computerized Mensural Music Editing Project (CMME) is conceived as an editorial initiative. The inner encoding (XML-based) is aimed to capture most details of mensural notation (graphical), as well as metadata (critical apparatus). The output consists of interactive scores allowing the comparison between original and modern notation, as well as between score variants. Analysis is facilitated for a single composition and fully relegated to the human analyst. The database is accessed via metadata browsing only. Data export is allowed for individual compositions, but conversion to other standard formats is not provided. The result is a highly curated database, but barely repurposable beyond the system. For its part, the Josquin Research Project (JRP) is conceived as an analytical initiative. The inner encoding (Humdrum-based) is aimed to facilitate highly sophisticated automated analysis. The database is accessed via metadata and content-based browsing, and it is fully searchable through musical content. Large datasets can be inspected by means of visual summaries of the complete compositions, and annotated scores are produced according to the user´s choices. The complete dataset is exportable, and available in many standard formats. The result is a semi-curated database (hosting modern editions only), but highly analyzable and repurposable.