Search

In constructive music theory, such as Schenkerian analysis and the Generative Theory of Tonal Music (GTTM), the hierarchical importance of pitch events is conveniently represented by a tree structure. Although a tree is intuitive and visible, such a graphic representation cannot be treated in mathematical formalization. Especially in the GTTM, the conjunction height of two branches is often arbitrary, contrary to the notion of hierarchy. As even a tree is a kind of graph, and a graph is often represented by a matrix, we show the linear algebraic representation of a tree, specifying the conjunction heights. Thereafter, we explain the ‘reachability’ between pitch events (corresponding to information about reduction) by the multiplication of matrices. In addition we discuss multiplication with vectors representing a sequence of harmonic functions, and suggest the notion of stability. Finally, we discuss operations between matrices with the objective of modelling compositional processes with simple algebraic operations.

This book constitutes the thoroughly refereed post-conference of the 11th International Symposium on Computer Music Modeling and Retrieval, CMMR 2015, held in Plymouth, UK, in June 2015.

This paper takes a step back from what are labeled ‘problem solving’approaches to the psychology of music memory and processing. In contrastwith generalised expectation theories of music processing, a hypothesis isproposed which is based on the idea that mental representations of music arelargely based on a large library of individual pieces: Case-based memory. Weargue that it is the activation of these memories that forms the critical aspect ofmusical experience. Furthermore, a specific hypothesis is proposed whichargues that it may be possible to represent any new piece of music through thechaining together of different, pre-existing veridical segments of music, incontrast with ‘problem solving by generalization’ which determines expectationbased on statistical/stylistic/schematic factors. By adjusting segment length, orby forming new segments through repeated listenings, new music can beabsorbed into an existing, growing mental database by chaining togetherexisting veridical segments that match the incoming stimulus.

This paper reports a “musical Turing test” conducted at a live concert ofalgorithm-generated performances, where one group of participants were invited torank the most human-like performance while knowing that one of the performanceswas by a human, and another group of participant were asked to do the same,but without knowing that there was a human performer on the program. Theprogram consisted of five pieces from the classical/romantic period, played on aDisklavier. High quality music-expression algorithms were used to generate thealgorithmic renditions. Regardless of the group, musical experience and a numberof other factors, the subjects were unable to identify the human performer out ofthe five performances. The group that did not know there was a human performerhad a wider range of votes compared to the group that did know. Furthermore,subjects were less confident of their answers when they knew that they werecomparing human and computer-generated performances . On the contrary, ifsubjects believed they were only comparing computer-generated performances thetask may have been less demanding. Findings suggest that computer algorithmsare able to substitute for human performance, but the role of the physical presenceof the performer (who was absent in this study) could be an area for furtherinvestigation.

This paper describes the interface design of our AeSon (Aesthetic Sonification) Toolkit motivated by user-centred customisation of the aesthetic representation and scope of the data. The interface design is developed from 3 premises that distinguish our approach from more ubiquitous sonification methodologies. Firstly, we prioritise interaction both from the perspective of changing scale, scope and presentation of the data and the user's ability to reconfigure spatial panning, modality, pitch distribution, critical thresholds and granularity of data examined. The user, for the majority of parameters, determines their own listening experience for real-time data sonification, even to the extent that the interface can be used for live data-driven performance, as well as traditional information analysis and examination. Secondly, we have explored the theories of Tufte, Fry and other visualization and information design experts to find ways in which principles that are successful in the field of information visualization may be translated to the domain of sonification. Thirdly, we prioritise aesthetic variables and controls in the interface, derived from musical practice, aesthetics in information design and responses to experimental user evaluations to inform the design of the sounds and display. In addition to using notions of meter, beat, key or modality and emphasis drawn from music, we draw on our experiments that evaluated the effects of spatial separation in multivariate data presentations.