"Compositional Approaches to New Media Paradigms" is the discursive accompaniment to the original composition BoMoH, (a new media chamber opera. A variety of new media concepts and practices are discussed in relation to their use as a contemporary compositional methodology for computer musicians and digital content producers. This paper aligns relevant discourse with a variety of concepts as they influence and affect the compositional process. This paper does not propose a new working method; rather it draws attention to a contemporary interdisciplinary practice that facilitates new possibilities for engagement and aesthetics in digital art/music. Finally, in demonstrating a selection of the design principals, from a variety of new media theories of interest, in compositional structure and concept, it is my hope to provide composers and computer musicians with a tested resource that will function as a helpful set of working guidelines for producing new media enabled art, sonic or otherwise.

Since the development of interactive music software in the 1980s, a new genre of works for clarinet and computer has emerged. The rapid proliferation of interactive music resulted in a great deal of experimentation, creating a lack of standardization in both the composition and performance of this repertoire. In addition, many performers are reluctant to approach these works due to unfamiliarity with the genre and its technical and musical considerations. Performance practice commonly refers to interpretation of a written score, but the technology involved in interactive music requires a broader definition of performance practice; one that also addresses computer software, coordination between the performer and computer system, and technology such as microphones and pedals. The problems and potential solutions of interactive music performance practice are explored in this paper through review of the relevant published literature, interviews with experts in the field, and examination of musical examples from works for clarinet and computer by Lippe, May, Pinkston, Rowe, and Welch. Performance practice considerations of interactive music fall into the categories of notation, technology, collaboration, interpretation, and rehearsal. From the interviews and the literature, it is clear that the performance of interactive music requires specific knowledge and skills that performers may not encounter in other genres of contemporary music, including microphone technique, spatialization, sound processing, and improvisation. Performance practice issues are often mediated by close collaboration between performers and composers, but they can inhibit the accessibility of these works to new performers, and may be detrimental to the long-term viability of interactive music. Recommendations for resolving these issues are directed at both composers and performers of interactive music. A listing of over one hundred interactive works for clarinet and computer is also included.

Examines the use of vernacular, country guitar styles in an electro-acoustic environment. Special attention is given to performance practices and explanation of techniques. Electro-acoustic techniques-including sound design and spatialization-are given with sonogram analyses and excerpts from the score. Compositional considerations are contrasted with those of Mario Davidovsky and Jean-Claude Risset with special emphasis on electro-acoustic approaches. Contextualization of the piece in reference to other contemporary, electric guitar music is shown with reference to George Crumb and Chiel Meijering.

No description available.

Computer-generated music composition programs have yet to produce creative, natural sounding music. To date, most approaches constrain the search space heuristically while ignoring the inherent structure of music over time. To address this problem, this thesis introduces NEAT Drummer, which evolves a special kind of artificial neural network (ANN) called compositional pattern producing networks (CPPNs) with the NeuroEvolution of Augmenting Topologies (NEAT) method for evolving increasingly complex structures. CPPNs in NEAT Drummer input existing human compositions and output an accompanying drum track. The existing musical parts form a scaffold i.e. support structure, for the drum pattern outputs, thereby exploiting the functional relationship of drums to musical parts (e.g. to lead guitar, bru:is, etc.) The results are convincing drum patterns that follow the contours of the original song, validating a new approach to computergenerated music composition.

Les récentes évolutions des technologies de l'information et de la communication font qu'il est aujourd'hui facile de consulter des catalogues de morceaux de musique conséquents. De nouvelles représentations et de nouveaux algorithmes doivent de ce fait être développés afin de disposer d'une vision représentative de ces catalogues et de naviguer avec agilité dans leurs contenus. Ceci nécessite une caractérisation efficace des morceaux de musique par l'intermédiaire de descriptions macroscopiques pertinentes. Dans cette thèse, nous nous focalisons sur l'estimation de la structure des morceaux de musique : il s'agit de produire pour chaque morceau une description de son organisation par une séquence de quelques dizaines de segments structurels, définis par leurs frontières (un instant de début et un instant de fin) et par une étiquette représentant leur contenu sonore.La notion de structure musicale peut correspondre à de multiples acceptions selon les propriétés musicales choisies et l'échelle temporelle considérée. Nous introduisons le concept de structure “sémiotique" qui permet de définir une méthodologie d'annotation couvrant un vaste ensemble de styles musicaux. La détermination des segments structurels est fondée sur l'analyse des similarités entre segments au sein du morceau, sur la cohérence de leur organisation interne (modèle “système-contraste") et sur les relations contextuelles qu'ils entretiennent les uns avec les autres. Un corpus de 383 morceaux a été annoté selon cette méthodologie et mis à disposition de la communauté scientifique.En termes de contributions algorithmiques, cette thèse se concentre en premier lieu sur l'estimation des frontières structurelles, en formulant le processus de segmentation comme l'optimisation d'un coût composé de deux termes~: le premier correspond à la caractérisation des segments structurels par des critères audio et le second reflète la régularité de la structure obtenue en référence à une “pulsation structurelle". Dans le cadre de cette formulation, nous comparons plusieurs contraintes de régularité et nous étudions la combinaison de critères audio par fusion. L'estimation des étiquettes structurelles est pour sa part abordée sous l'angle d'un processus de sélection d'automates à états finis : nous proposons un critère auto-adaptatif de sélection de modèles probabilistes que nous appliquons à une description du contenu tonal. Nous présentons également une méthode d'étiquetage des segments dérivée du modèle système-contraste.Nous évaluons différents systèmes d'estimation automatique de structure musicale basés sur ces approches dans le cadre de campagnes d'évaluation nationales et internationales (Quaero, MIREX), et nous complétons cette étude par quelques éléments de diagnostic additionnels. / Recent progress in information and communication technologies makes it easier to access large collections of digitized music. New representations and algorithms must be developed in order to get a representative overview of these collections, and to browse their content efficiently. It is therefore necessary to characterize music pieces through relevant macroscopic descriptions. In this thesis, we focus on the estimation of the structure of music pieces : the goal is to produce for each piece a description of its organization by means of a sequence of a few dozen structural segments, each of them defined by its boundaries (starting time and ending time) and a label reflecting its audio content.The notion of music structure corresponds to a wide range of meanings depending on the musical properties and the temporal scale under consideration. We introduce an annotation methodology based on the concept of “semiotic structure" which covers a large variety of musical styles. Structural segments are determined through the analysis of their similarities within the music piece, the coherence of their inner organization (“system-contrast" model) and their contextual relationship. A corpus of 383 pieces has been annotated according to this methodology and released to the scientific community.In terms of algorithmic contributions, this thesis concentrates in the first place on the estimation of structural boundaries. We formulate the segmentation process as the optimization of a cost function which is composed of two terms. The first one corresponds to the characterization of structural segments by means of audio criteria. The second one relies on the regularity of the target structure with respect to a “structural pulsation period". In this context, we compare several regularity constraints and study the combination of audio criteria through fusion.Secondly, we consider the estimation of structural labels as a probabilistic finite-state automaton selection process : in this scope, we propose an auto-adaptive criterion for model selection, applied to a description of the tonal content. We also propose a labeling method derived from the system-contrast model.We evaluate several systems for structural segmentation of music based on these approaches in the context of national and international evaluation campaigns (Quaero, MIREX). Additional diagnostic is finally presented to complement this work.

La coexistence de deux formats : cinéma à 24 images/s et vidéo à<br />25 images/s, implique l'accélération ou le ralentissement de la<br />bande-son lors du transfert d'un format vers l'autre. Ceci<br />provoque une modification temporelle du signal sonore, et par<br />conséquent une modification spectrale avec altération du timbre.<br />Les studios de post-production audiovisuelle souhaitent compenser<br />cet effet par l'application d'une transformation sonore adéquate.<br /><br />L'objectif de ce travail est de fournir à l'industrie<br />audiovisuelle un système permettant de pallier la modification de<br />timbre engendrée par le changement de vitesse de lecture. Ce<br />système se compose d'une part d'un algorithme de traitement et<br />d'autre part d'une machine sur lequel il est implanté.<br />L'algorithme est conçu et développé pour répondre aux contraintes<br />liées à la qualité sonore et à la compatibilité multicanal. La<br />machine, baptisée HARMO, est conçue spécifiquement par la société<br />GENESIS sur la base de processeurs de signaux numériques, et doit<br />répondre à la contrainte de temps-réel. Cet aspect "valorisation"<br />conduit à intégrer dans le projet les contraintes de coût et de<br />délai de réalisation.<br /><br />Un état de l'art basé sur une bibliographie quasi-exhaustive<br />aboutit à une classification originale des méthodes de dilatation<br />et de transposition existantes. Ceci nous amène à distinguer et à<br />étudier les méthodes classiques temporelles et fréquentielles, et<br />à introduire les méthodes temps-fréquence. Cette classification<br />est à la base de plusieurs méthodes innovantes :<br /><br />1. deux méthodes temps-fréquence dont l'analyse est adaptée à l'audition,<br /><br />2. deux méthodes couplées qui associent les avantages des méthodes temporelles et fréquentielles,<br /><br />3. une méthode temporelle basée sur une amélioration des méthodes existantes.<br /><br />Les algorithmes sont évalués grâce à une banque de sons-test<br />spécifiquement élaborée pour mettre en évidence les défauts<br />caractéristiques des algorithmes. Notre choix final s'est porté<br />sur l'approche temporelle, que nous optimisons par l'adjonction de<br />critères de segmentation basés sur l'autocorrélation normalisée et<br />la détection de transitoires. Cet algorithme s'intègre dans un<br />logiciel qui a été structuré pour un fonctionnement temps-réel et<br />multicanal sur le système HARMO.

Note sequence morphing is the combination of two note sequences to create a ‘hybrid transition’, or ‘morph’. The morph is a ‘hybrid’ in the sense that it exhibits properties of both sequences. The morph is also a ‘transition’, in that it can segue between them. An automated and interactive approach allows manipulation in realtime by users who may control the relative influence of source or target and the transition length. The techniques that were developed through this research were designed particularly for popular genres of predominantly instrumental electronic music which I will refer to collectively as Mainstream Electronic Music (MEM). The research has potential for application within contexts such as computer games, multimedia, live electronic music, interactive installations and accessible music or “music therapy”. Musical themes in computer games and multimedia can morph adaptively in response to parameters in realtime. Morphing can be used by electronic music producers as an alternative to mixing in live performance. Interactive installations and accessible music devices can utilise morphing algorithms to enable expressive control over the music through simple interface components. I have developed a software application called LEMorpheus which consists of software infrastructure for morphing and three alternative note sequence morphing algorithms: parametric morphing, probabilistic morphing and evolutionary morphing. Parametric morphing involves converting the source and target into continuous envelopes, interpolation, and converting the interpolated envelopes back into note sequences. Probabilistic morphing involves converting the source and target into probability matrices and seeding them on recent output to generate the next note. Evolutionary morphing involves iteratively mutating the source into multiple possible candidates and selecting those which are judged as more similar to the target, until the target is reached. I formally evaluated the probabilistic morphing algorithm by extracting qualitative feedback from participants in a live electronic music situation, benchmarked against a live, professional DJ. The probabilistic algorithm was competitive, being favoured particularly for long morphs. The evolutionary morphing algorithm was formally evaluated using an online questionnaire, benchmarked against a human composer/producer. For particular samples, the morphing algorithm was competitive and occasionally seen as innovative; however, the morphs created by the human composer typically received more positive feedback, due to coherent, large scale structural changes, as opposed to the forced continuity of the morphing software.

Μία από τις βασικές προκλήσεις στο μουσικό αυτοσχεδιασμό είναι ο διαδραστικός αυτοσχεδιασμός μεταξύ ενός ανθρώπου και ενός συστήματος. Στη παρούσα ενότητα παρουσιάζουμε ένα μουσικό διαδραστικό σύστημα (Πολύμνια) ως συνεχιστή της μελωδίας (as melody continuator). Για κάθε μουσικό πρότυπο (pattern) που έχει δοθεί από το χρήστη, το ευφυές σύστημα ανακαλεί ένα όμοιο (similar) γενικό πρότυπο που είναι αποθηκευμένο στη βάση του (database) και το οποίο το αναμορφώνει ανάλογα (reform). Το προτεινόμενο σύστημα κατευθύνει τη μουσική αναπαράσταση και την ομοιότητα του μουσικού προτύπου (musical pattern similarity) στη χρήση της εξόρυξης δεδομένων (data mining). Προτείνουμε ένα σχήμα μουσικής αναπαράστασης το οποίο μπορεί να χρησιμοποιηθεί για ανάλυση εξόρυξης δεδομένων (data mining analysis) η οποία στοχεύει στη μάθηση γενικών προτύπων και για τη συχνότητα και για τη διάρκεια σε συγκεκριμένα είδη μουσικής (music styles). Η εξόρυξη δεδομένων είναι μια αναδυόμενη διαδικασία μηχανικής μάθησης με την εξαγωγή προηγουμένως άγνωστων, αγώγιμων (actionable) πληροφοριών από πολύ μεγάλες επιστημονικές και εμπορικές βάσεις δεδομένων. Η μηχανική μάθηση (machine learning) έχει παίξει έναν κρίσιμο ρόλο στη υπολογιστική μουσική (computer music) σχεδόν από την αρχή της. Πρόσφατα η έρευνα στο πεδίο έχει εστιαστεί στην εξόρυξη μουσικής (music mining). Παρουσιάζουμε επίσης πειραματικά αποτελέσματα για έλεγχο και αξιολόγηση της αποδοτικότητας (efficiency) και της ακρίβειας του προτεινόμενου συστήματος «Πολύμνια». / One of the main challenges in music improvisation is interactive improvisation between a human and a system. In this thesis we present a musical interactive system (called polyhymnia) acting as melody continuator. For each musical pattern given by the user, it recalls a similar general pattern stored in its memory and reforms it. The proposed system addresses music representation and musical pattern similarity using data mining. We propose a scheme for monophonic music representation as traditional data sets suitable for common data mining algorithms and investigate the application of clustering similarity measures to musical pattern similarity. Data Mining is an emerging machine learning process of extracting previously unknown, actionable information from very large scientific and commercial databases. Machine learning has played a crucial role in the computer music almost since its beginning. Recently, research in the field has focused on music mining. We also present experimental results for testing and evaluating the efficiency and accuracy of the proposed system “polyhymnia”.

In the summer of 1953, Karlheinz Stockhausen began composing his first piece of elektronische Musik at the Westdeutscher Rundfunk Studio for Electronic Music in Cologne. Up to that point, Stockhausen's only experience with electroacoustic music was his time spent at the Radiodiffusion-Télévision Française the previous year, where he assisted Pierre Schaeffer and composed a piece of musique concrète. An early case study in the marriage of serial aesthetics and electroacoustic techniques, Studie I is a rigorously organized work that reflects Stockhausen's compositional philosophy of a unified structural principle in which all musical materials and parametric values are generated by and arranged according to a single governing series. In spite of this meticulously wrought serial structure, Studie I displays features that are the consequences of the realities of electronic sound production either imposing on the sonic result, or altering the compositional plan entirely. I use a three-part approach to my analysis of Studie I by examining Stockhausen's serial system, the electroacoustic studio techniques in use in 1953, and the original recorded realization through spectrographic analysis. Using this methodology, I expose the blurring of the supposed divide between elektronische Musik and musique concrète by exploring the features that lie between the serial plan and the technical processes Stockhausen used to realize Studie I.