Thesis (M.F.A.)--Ohio State University, 2008. / Title from first page of PDF file. Includes bibliographical references (p. 93-96).

Recently, there have been efforts to design more efficient ways to internalize music by applying the disciplines of cognition, psychology, temporality, aesthetics, and philosophy. Bringing together the fields of art and science, computational techniques can also be applied to musical analysis. Although a wide range of research projects have been conducted, the automatization of music analysis remains emergent. Importantly, patterns are revealed by using automated tools to analyze core musical elements created from melodies, harmonies, and rhythms, high-level features that are perceivable by the human ear. For music to be captured and successfully analyzed by a computer, however, one needs to extract certain information found in the lower-level features of amplitude, frequency, and duration. Moreover, while the identity of harmonic progressions, melodic contour, musical patterns, and pitch quantification are crucial factors in traditional music analysis, these alone are not exclusive. Visual representations are useful tools that reflect form and structure of non-conventional musical repertoire. Because I regard the fluidity of music and visual shape as strongly interactive, the ultimate goal of this thesis is to construct a practical tool that prepares the visual material used for musical composition. By utilizing concepts of time, computation, and composition, this tool effectively integrates computer science, signal processing, and music perception. This will be obtained by presenting two concepts, one abstract and one mathematical, that will provide materials leading to the original composition. To extract the desired visualization, I propose a fully automated tool for musical analysis that is grounded in both the mid-level elements of loudness, density, and range, and low-level features of frequency and duration. As evidenced by my sinfonietta, Equilibrium, this tool, capable of rapidly analyzing a variety of musical examples such as instrumental repertoire, electro-acoustic music, improvisation and folk music, is highly beneficial to my proposed compositional procedure.

The purpose of this study was to create a new method for visualizing music. Although many music visualizations already exist, this research was focused on creating high-quality, high-complexity animations that cannot be matched by real-time systems. There should be an obvious similarity between the input music and the final animation, based on the music information which the user decides to extract and visualize. This project includes a pipeline for music data extraction and creation of an editable visualization file. Within the pipeline, a music file is read into a custom analysis tool and time-based data is extracted. This data is output and then read into Autodesk Maya. The user may then manipulate the visualization as they see fit using the tools within Maya and render out a final animation. The default result of this process is a Maya scene file which makes use of the dynamics systems available to warp and contort a jelly-like cube. A variety of other visualizations may be obtained by mapping the data to different object attributes within the Maya interface. When rendered out and overlaid onto the music, there was a recognizable correlation between elements in the music and the animations in the video. This study shows that an accurate musical visualization may be achieved using this pipeline. Also, any number of different music visualizations may be obtained with relative ease when compared to the manual analysis of a music file or the manual animation of Maya objects to match elements in the music.

Italian Polka is an experiment that builds a bridge between Music and the field of Costume Design. It explores the new relationship of integration and artistic possibility between Music, Costume Design, Dance, and Digital Art. This is also an attempt to participate in a new form of performing art: a combination of a live concert and a costume show. / text

Os layouts gerados por técnicas de projeção multidimensional podem ser a base para diferentes metáforas de visualização que são aplicáveis a diversos tipos de dados. Existe muito interesse em investigar metáforas alternativas à comumente usada, nuvem de pontos usada para exibir layouts gerados por projeções multidimensionais. Neste trabalho, foi estudado este problema, com foco no domínio da visualização de músicas. Existem muitas dimensões envolvidas na percepção e manipulação de músicas e portanto é difícil encontrar um modelo computacional intuitivo para representá-las. Nosso objetivo neste trabalho foi investigar as representações visuais capazes de transmitir a estrutura de uma música, assim como exibir uma coleção de músicas de modo a ressaltar as similaridades. A solução proposta consiste em uma representação icônica de músicas individuais, que é associada ao posicionamento espacial dos grupos ou coleções de músicas gerado por uma técnica de projeção multidimensional que reflete suas similaridades estruturais. Tanto a projeção quanto o ícone requerem um vetor de características para representar a música. As características são extraídas a partir de arquivos MIDI, já que a própria natureza das descrições MIDI permite a identificação das estruturas musicais relevantes. Estas características proporcionam a entrada tanto para a comparação de dissimilaridades quanto para a construção do ícone da música. Os posicionamentos espaciais são obtidos usando a técnica de projeção multidimensional Least Square Projection (LSP), e as similaridades são calculadas usando a distância Dynamic Time Warping (DTW). O ícone fornece um resumo visual das repetições de acordes em uma música em particular. Nessa dissertação são descritos os processos de geração destas representações visuais, além de descrever um sistema que implementa esses recursos e ilustrar como eles podem apoiar algumas tarefas exploratórias das coleções de músicas, identificando possíveis cenários de uso / The layouts generated by multidimensional projection techniques can be the basis for different visualization metaphors that are applicable to various data types. There is much interest in investigating alternatives to the point cloud metaphor commonly used to present projection layouts. In this work, we investigated this problem, targeting the domain of music visualization. There are many dimensions involved in the perception and manipulation of music and therefore it is difficult to find an intuitive computer model to represent music. Our goal in this work was to investigate visual representations capable of conveying the musical structure of a song, as well as displaying a collection of songs so as to highlight their similarities. The proposed solution consists of an iconic representation for individual songs, that is associated with the spatial positioning of groups or collections of songs generated by a multidimensional projection technique that reflects their structural similarity. Both the projection and the icon require a feature vector representation of the music. The features are extracted from MIDI files, as the nature of the MIDI descriptions allows the identification of the relevant musical structures. These features provide the input for both the dissimilarity comparison and for constructing the music icon. The spatial layout is computed with the Least Square Projection (LSP) technique, and similarities are computed using the Dynamic Time Warping (DTW) distance. The icon provides a visual summary of the chord repetitions in a particular song. We describe the process of generating these visual representations, describe a system that implements such funcionalities and illustrate how they can support some exploratory tasks on music collections, identifying possible usage scenarios

Os layouts gerados por técnicas de projeção multidimensional podem ser a base para diferentes metáforas de visualização que são aplicáveis a diversos tipos de dados. Existe muito interesse em investigar metáforas alternativas à comumente usada, nuvem de pontos usada para exibir layouts gerados por projeções multidimensionais. Neste trabalho, foi estudado este problema, com foco no domínio da visualização de músicas. Existem muitas dimensões envolvidas na percepção e manipulação de músicas e portanto é difícil encontrar um modelo computacional intuitivo para representá-las. Nosso objetivo neste trabalho foi investigar as representações visuais capazes de transmitir a estrutura de uma música, assim como exibir uma coleção de músicas de modo a ressaltar as similaridades. A solução proposta consiste em uma representação icônica de músicas individuais, que é associada ao posicionamento espacial dos grupos ou coleções de músicas gerado por uma técnica de projeção multidimensional que reflete suas similaridades estruturais. Tanto a projeção quanto o ícone requerem um vetor de características para representar a música. As características são extraídas a partir de arquivos MIDI, já que a própria natureza das descrições MIDI permite a identificação das estruturas musicais relevantes. Estas características proporcionam a entrada tanto para a comparação de dissimilaridades quanto para a construção do ícone da música. Os posicionamentos espaciais são obtidos usando a técnica de projeção multidimensional Least Square Projection (LSP), e as similaridades são calculadas usando a distância Dynamic Time Warping (DTW). O ícone fornece um resumo visual das repetições de acordes em uma música em particular. Nessa dissertação são descritos os processos de geração destas representações visuais, além de descrever um sistema que implementa esses recursos e ilustrar como eles podem apoiar algumas tarefas exploratórias das coleções de músicas, identificando possíveis cenários de uso / The layouts generated by multidimensional projection techniques can be the basis for different visualization metaphors that are applicable to various data types. There is much interest in investigating alternatives to the point cloud metaphor commonly used to present projection layouts. In this work, we investigated this problem, targeting the domain of music visualization. There are many dimensions involved in the perception and manipulation of music and therefore it is difficult to find an intuitive computer model to represent music. Our goal in this work was to investigate visual representations capable of conveying the musical structure of a song, as well as displaying a collection of songs so as to highlight their similarities. The proposed solution consists of an iconic representation for individual songs, that is associated with the spatial positioning of groups or collections of songs generated by a multidimensional projection technique that reflects their structural similarity. Both the projection and the icon require a feature vector representation of the music. The features are extracted from MIDI files, as the nature of the MIDI descriptions allows the identification of the relevant musical structures. These features provide the input for both the dissimilarity comparison and for constructing the music icon. The spatial layout is computed with the Least Square Projection (LSP) technique, and similarities are computed using the Dynamic Time Warping (DTW) distance. The icon provides a visual summary of the chord repetitions in a particular song. We describe the process of generating these visual representations, describe a system that implements such funcionalities and illustrate how they can support some exploratory tasks on music collections, identifying possible usage scenarios

Neste trabalho é apresentada a biblioteca AIM, uma biblioteca de objetos Pd que combina ferramentas de análise de áudio e síntese de imagens para a geração de acompanhamentos visuais para uma entrada musical. O usuário estabelece conexões que determinam como os parâmetros musicais afetam a síntese de objetos gráficos, e controla estas conexões em tempo-real durante a performance. A biblioteca combina um protocolo de comunicação para intercâmbio de parâmetros musicais e visuais com uma interface fácil de usar, tornando-a acessível a usuários sem experiência em programação de computadores. Suas áreas possíveis de aplicação incluem a musicalização infantil e a indústria de entretenimento. / This thesis describes the AIM library, a Pd object library which combines audio analysis and image synthesis tools for generating visual accompaniments to musical input data. The user establishes connections that determine how musical parameters affect the synthesis of graphical objects, and controls these connections in real-time during performance. The library combines a straightforward communication protocol for exchanging musical and visual parameters with an easy-to-use interface that makes it accessible for users with no computer programming experience. Its potential applications areas include children\'s musical education and the entertainment industry.

This dissertation features methods of analyzing symbolic music, focused on n-gram-based approaches, as this representation resembles the most text and natural languages. The analysis of similarities between several text and music corpora is accompanied with implementation of text-based methods for problems of composer classification and symbolic music similarity definition. Both problems contain thorough evaluation of performance of the systems with comparisons to other approaches on existing testbeds. It is also described how one can use this symbolic representation in conjunction with genetic algorithms to tackle problems like melody generation. The proposed method is fully automated, and the process utilizes n-gram statistics from a sample corpus to achieve it. A method of visualization of complex symbolic music pieces is also presented. It consist of creating a self similarity matrix of a piece in question, revealing dependencies between voices, themes and sections, as well as music structure. A fully automatic technique of inferring music structure from these similarity matrices is also presented The proposed structure analysis system is compared against similar approaches that operate on audio data. The evaluation shows that the presented structure analysis system outperformed significantly all audio-based algorithms available for comparison in both precision and recall.

Neste trabalho é apresentada a biblioteca AIM, uma biblioteca de objetos Pd que combina ferramentas de análise de áudio e síntese de imagens para a geração de acompanhamentos visuais para uma entrada musical. O usuário estabelece conexões que determinam como os parâmetros musicais afetam a síntese de objetos gráficos, e controla estas conexões em tempo-real durante a performance. A biblioteca combina um protocolo de comunicação para intercâmbio de parâmetros musicais e visuais com uma interface fácil de usar, tornando-a acessível a usuários sem experiência em programação de computadores. Suas áreas possíveis de aplicação incluem a musicalização infantil e a indústria de entretenimento. / This thesis describes the AIM library, a Pd object library which combines audio analysis and image synthesis tools for generating visual accompaniments to musical input data. The user establishes connections that determine how musical parameters affect the synthesis of graphical objects, and controls these connections in real-time during performance. The library combines a straightforward communication protocol for exchanging musical and visual parameters with an easy-to-use interface that makes it accessible for users with no computer programming experience. Its potential applications areas include children\'s musical education and the entertainment industry.

No description available.