Dig that Lick. Analysing large-scale data for melodic patterns in jazz performances

点击翻译按钮获取中文摘要

The recorded legacy of jazz performance now spans a century and offers a trove of knowledge about improvisation for exploration and analysis. A thorough account of these practices involved and processes in which jazz is performed is now possible thanks to significant advances in the fields of computational analysis of audio content (including Music Information Retrieval) and information management technologies (including Big Data and Semantic Web technologies). The former enables the automatic description of audio recordings in terms of high-level musical aspects, and the latter allows such analyses to be linked to bibliographic metadata describing performers, composers, and arrangers as well as historical and social metadata describing listeners and contexts of music performance and consumption. These technologies are now reaching a level where automatic semantic metadata extraction and aggregation technologies can facilitate access to large collections in an unprecedented way, fostering new interdisciplinary research opportunities. In a close collaboration between musicologists and computer scientists we will develop the methods to create a deeper and more comprehensive understanding of jazz music and its evolution in relation to the social and cultural context.The project focuses on the development and history of melodic patterns (or "licks") in jazz improvisations. As a purposeful effort, we will complete the full cycle of methods from raw media data (audio recordings) to an understanding of the transmission and transformation of patterns, which will be aesthetically contextualised and historically situated. The project brings together renowned scholars and results from several high-profile projects on both sides of the Atlantic to assemble the technical infrastructure, create the tools, conduct the technical analysis and develop the humanistic and social scientific methods for this domain and cultural studies more generally.The project seeks to extend the research paradigms of jazz studies by:- optimising existing infrastructures for the deployment of semantic audio analyses over large collections;- investigating the benefits and barriers of state-of-the art Semantic Web technologies for information management;- facilitating access to large collections of audio and associated metadata by creating easy to use interfaces for content selection, semantic analysis, and aggregation of;- using the developed infrastructure to analyse the use of melodic patterns in a large jazz corpus; and- relating the analytic results to metadata and background knowledge representations in order to trace and interpret musical influence across time and space as well as various historical and social context.