Sound and Music Computing

This article provides details on the implementation and design of the 'Make Your Own Band' software suite, a suite of standalone software applications utilizing machine learning (ML) and music information retrieval (MIR) developed for use in school classrooms. Preliminary results through testing, surveying, and interviewing music teachers suggest that the system holds significant promise for advancing music education. It could promote inclusion through kinesthetic and gestural methods, has the potential to simplify music activities, address instrument shortages in resourcelimited environments, enhance participation, and increase student engagement.

The most common interface for detailed information communication by computer is the QWERTY keyboard. However users have been shown to be notoriously bad at communicating their emotional state when typing emails or SMS messageshence the reason for so many misunderstood messages, and for netiquette and emoticons. In studies of expressive performance, it has been shown that there are consistencies in the playing styles of pianists and the emotion they are trying to express. There has been initial work on algorithms to emulate such emotional playing styles on computers, and to detect emotion playing styles from MIDI transcripts. These include such elements as tempo and loudness. The emotional playing styles are thought to be related toamongst other things -the performer's emotional state and the biophysical limitations involved in utilizing a piano keyboard. We propose examining whether such performance emotion detection algorithms can be utilized in examining the emotional state in real-time of a typist using a QWERTY keyboard. Such data would be useful in Human-Computer Interaction studies, and automated emoticon generation systemsamongst others. It also leads naturally to proposals for a protocol for emotional ASCII transmission based on the MIDI protocol.

This paper introduces a system that uses brainwaves, or EEG (electroencephalogram), information to steer generative rules in order to compose and perform music on the fly. The paper starts by noting the various attempts at the design of systems to produce music from EEG, followed by a short technical introduction to EEG sensing and analysis. Next, it introduces the generative component of the system, which employs Artificial Intelligence techniques (e.g., ATN grammars) for computer-replication of musical styles. Then, it presents a demonstration system that constantly monitors the EEG of the subject and activates generative rules that are associated with the most prominent frequency band in the spectrum of the EEG signal. The system also measures the complexity of the EEG signal in order to modulate the tempo (beat) and dynamics (loudness) of the performance. The paper concludes with a brief discussion on the achievements and limitations of our research so far, and comments on its contribution for the development of assistive technology for severe physical and neurological disability, which is one of the main goals of this work.

Research evaluating perceptual responses to music has identified many structural features as correlates that might be incorporated in computer music systems for affectively charged algorithmic composition and/or expressive music performance. In order to investigate the possible integration of isolated musical features to such a system, a discrete feature known to correlate some with emotional responses -rhythmic density -was selected from a literature review and incorporated into a prototype system. This system produces variation in rhythm density via a transformative process. A stimulus set created using this system was then subjected to a perceptual evaluation. Pairwise comparisons were used to scale differences between 48 stimuli. Listener responses were analysed with Multidimensional scaling (MDS). The 2-Dimensional solution was then rotated to place the stimuli with the largest range of variation across the horizontal plane. Stimuli with variation in rhythmic density were placed further from the source material than stimuli that were generated by random permutation. This, combined with the striking similarity between the MDS scaling and that of the 2-dimensional emotional model used by some affective algorithmic composition systems, suggests that isolated musical feature manipulation can now be used to parametrically control affectively charged automated composition in a larger system.

This paper introduces a new brain-computer interface (BCI) system that uses electroencephalogram (EEG) information to steer generative rules to compose and perform music. It starts by noting the various attempts at the design of BCI systems, including systems for music. Then it presents a short technical introduction to EEG sensing and analysis. Next, it introduces the generative music component of the system, which employs our own adapted version of a machine-learning technique based on ATNs (Augmented Transition Networks) for the computer-replication of musical styles. The system constantly monitors the EEG of the subject and activates generative rules associated with the activity of different frequency bands of the spectrum of the EEG signal. The system also measures the complexity of the EEG signal to modulate the tempo (beat) and dynamics (loudness) of the performance. Subjects can be trained to select between different classes of generative rules to produce original pieces of music.

Artificial Life (A-Life) and Evolutionary Algorithms (EA) provide a variety of new techniques for making and studying music. EA have been used in different musical applications, ranging from new systems for composition and performance, to models for studying musical evolution in artificial societies. This paper starts with a brief introduction to three main fields of application of EA in Music, namely sound design, creativity and computational musicology. Then it presents our work in the field of computational musicology. Computational musicology is broadly defined as the study of Music with computational modelling and simulation. We are interested in developing A-Life-based models to study the evolution of musical cognition in an artificial society of agents. In this paper we present the main components of a model that we are developing to study the evolution of musical ontogenies, focusing on the evolution of rhythms and emotional systems. The paper concludes by suggesting that A-Life and EA provide a powerful paradigm for computational musicology.

This paper presents a system that implements a society of agent performers that evolve expressive music performances (EMP) through their interactions. Each agent performer evaluates a performance using a fitness function derived from the structure of the performing piece. Agents are born with different fitness functions, representing preferences for different types of performance. A new-born agent firstly applies a Genetic Algorithm to generate a refined performance with its fitness function. Through an agent's life, it listens to the performances of others' that it connects to in the social network. Given specific conditions, it may try imitate some performances and modify its own preference. An agent performer dies of age or inactivity. New agent performers are occasionally created. With this model, we aim to generate expressive performances through simulating the effect of social pressure and culture transmission.

This paper draws together two strands of research in different disciplines: EEG Frontal Asymmetry studies and Affective-based Algorithmic Composition and Performance. There have been numerous studies which support the hypothesis that Frontal EEG asymmetry is an indicator of arousal and valence of emotion. There have also been a number of studies examining which music features are related to which emotional perception, in the generative composition of, and performance of, music. This paper ...

A multi-agent system is presented which generates melody pitch sequences with a hierarchical structure. The system has no explicit melodic intelligence and generates the pitches as a result of emotional influence and communication between agents, and the hierarchical structure is a result of the emerging agent social structure. Another key element is that the system is not a mapping from multi-agent interaction onto musical features, but actually utilizes music for the agents to communicate emotions. Each agent in the society learns its own growing tune during the interaction process.

We detail our work on a multi-agent model for the generation of structurally expressive music performance. It is a combined performance and composition system which attempts to implement the common observation in expressive music performance studies that performers slow ...

We introduce the concept of Expressive Performance Generated Composition where the initial specification for a composition comes from the details of an expressive performance. This expressive performance may have been of a previous piece or be specially produced for the ...

This paper will present a survey of electronic music and sonic art applications of Cellular Automata in the MIDI domain, and of the implications of that work for future developments. Algorithmic and computational processes have been of interest to artists in the audio domain for many years, creating an emerging culture of algorithmic generative electronic art. There have been several approaches at applying Cellular Automata in the production of electronic music and sonic art. Creative domains of application exist in the ...

This study explores how listeners use verb constructions to describe everyday sounds. It considers (i) how the verb constructions provide Gestalt to events in the form of activities, processes and states, and (ii) how the constructions correspond to different aspects of a soundscape (sound sources, acoustic properties, and listener reactions). We found that verb constructions primarily describe sound sources and that these are portrayed as dynamic events. Acoustic properties were rarely described through verb constructions. When they were, the constructions primarily conveyed static properties of sound as in the sound is loud as opposed to dynamic events as in the sound comes closer. Listener reactions were primarily expressed through comparative constructions such as it sounds like and justifications such as I can hear. The upshot of the study is that speakers conceptualise sounds according to highly dynamic Gestalts for which the main focus is the source of the sound.

In this work, we introduce musif, a Python package that facilitates the automatic extraction of features from symbolic music scores. The package includes the implementation of a large number of features, which have been developed by a team of experts in musicology, music theory, statistics, and computer science. Additionally, the package allows for the easy creation of custom features using commonly available Python libraries. musif is primarily geared towards processing high-quality musicological data encoded in MusicXML format, but also supports other formats commonly used in music information retrieval tasks, including MIDI, MEI, Kern, and others. We provide comprehensive documentation and tutorials to aid in the extension of the framework and to facilitate the introduction of new and inexperienced users to its usage.

This paper introduces i-Berlioz, a proof-of-concept interactive Computer-Aided Orchestration (CAO) system that suggests combinations of musical instruments to produce timbres specified by the user by means of verbal descriptors. The system relies on machine learning for timbre classification and generative orchestration, and tools to write and execute interactive scenarios. The paper details the two main parts of i-Berlioz: its generative and exploratory engines, respectively, accompanied with examples. The authors discuss how i-Berlioz can aid the composition of musical form based on timbre.

A proposito della costituzione dello Studio di fonologia della RAI di Milano, ricorda Bruno Maderna: "S'era dovuto convincere una quantità di gente, s'erano prodotti, con i primi scarsi mezzi messici a disposizione numerevoli 'effetti', pezzi e pezzettini ad usun Delfini, roba a buon mercato ma che piaceva ai palati non troppo sottili dei funzionari e dei cosiddetti registi della televisione. Quasi di nascosto si era riusciti a produrre qualche lavoro di una certa serietà; ricordo ancora i primi lavori semplici ma faticosamente e con mille dubbi portati a compimento: "Mutazioni" di Berio e il mio "Notturno". Il mio primo lavoro puramente elettronico, composto di soli gruppi di frequenze, suono bianco, miscellanee varie, venne realizzato nottetempo".
Luciano Berio così ricorda quegli anni: "Nell'ottobre del 1952 ero a New York dove, al Museum of Modern Art, assistetti a un concerto diretto da Leopold Stokowski che, a metà del concerto, si rivolse al pubblico dicendo "adesso vi lascio, due altoparlanti prenderanno il mio posto". Fu in quell'occasione che ascoltai per la prima volta musiche elettroniche (la chiamavano tape music), erano di Vladimir Ussachevsky e di Otto Luening. [...] Questi due musicisti mi avevano raccontato come fabbricavano quei suoni e con quali apparecchiature. Arrivato a Milano, alla RAI, mi sono procurato dei filtri, degli oscillatori, dei generatori di suono bianco e mi sono fatto allestire un variatore di velocità: le stesse povere apparecchiature, cioè, che abbiamo impiegato per la realizzazione, con Maderna, di "Ritratto di città". [...] Tornato in Italia ho sentito parlare del lavoro di Meyer-Eppler, di Eimert e di Stockhausen e ne sono rimasto profondamente colpito".

This patent is based on a novel model of human categorial perception of harmonic speech sounds, alternative to the currently dominant formant theory and to the NLP procedures based on the concepts of amplitude envelope and Mel-frequency scale. The field of the present invention relates to recognition or synthesis of human-uttered speech. In particular, computer-implemented methods are disclosed for recognizing or synthesizing human- uttered harmonic sounds.

Si nous sommes habitués à topographier, photographier, ou décrire les grottes, quid de leur dimension sonore ? Pour pallier ce manque d'informations nous avons investi la grotte du Déroc à Vallon-Pont-d'Arc afin d'y appréhender la dimension acoustique. A partir d'observations à la voix et à l’oreille, nous avons repéré les zones de résonance ; soit sur une fréquence
isolée, soit sur une partie du spectre vocal, ou encore les zones où une réverbération du son est remarquable. Ensuite, pour étudier ces espaces acoustiques repérés, nous avons mesuré leur réponse en fréquence et leur temps de réverbération. Pour cela nous utiliserons la méthode de déconvolution, qui consiste à émettre et enregistrer un bruit bien choisi, puis à comparer le signal émis au signal reçu. Cela donne une identité acoustique propre à l'espace étudié. Dans notre cas, nous émettons tour à tour un son continu dont la fréquence varie de 20 à 20 000 Hz, et une impulsion de bruit blanc produite par l'éclatement d'un ballon de baudruche. Cette étude acoustique au Déroc nous amène à nous interroger sur l'approche archéoacoustique déjà réalisée par le passé.

In response to the proposal of digitizing the entire back-run of several European audio archives, many research projects have been carried out in order to discover the technical issues involved in making prestigious audio documents digitally available, which are related to the A/D transfer process and supervised metadata extraction. This article gives an innovative approach to metadata extraction from such a complex source material. This article also describes the protocols defined, the processes undertaken, the results ascertained from several audio documents preservation projects and the techniques used. In addition, a number of recommendations are given for the re-recording process, aimed at minimizing the information loss and to automatically measure the unintentional alterations introduced by the A/D equipment.

Background: Auditory alarms in medical environments are widely criticized due to problematic design issues affecting learnability, recognition, and annoyance. To overcome these issues, new alarm sets vary properties like pitch, timing, and timbre-the "quality" of a sound. Past work identifies timbre as a promising candidate for improving alarm recognition, however most studies manipulate multiple acoustic parameters, obfuscating the independent effect of timbre on melodic perception. Methods: To better understand timbre's potential for improving alarm recognizability and reducing annoyance, we recruited forty-four participants for a melody recognition experiment. In a learning phase, participants learned short alarm melodies performed with either a complex musical timbre based on a xylophone or one following the specifications of a universal alarm standard. During a recall phase, participants heard the melodies again presented in either the xylophone or standard timbre and attempted to identify the corresponding label for each alarm. Afterward, participants heard alarm pairs presented in both timbres and chose which one they perceived to be more annoying. Results: Participants recalled xylophone alarm melodies with comparable accuracy to the standard timbre, however they found the musical timbre substantially less annoying. Regardless of the timbre at exposure, recognition of familiar melodies decreased if the timbre changed between learning and recall. Conclusions: We compared the recognizability and annoyance of alarm melodies presented in either a standard alarm timbre or a synthesized xylophone timbre. Our findings reveal musical sounding alarms are comparably recognizable, yet significantly less annoying than alarm signals common in medical environments. These outcomes provide a promising first step to improving patient care through musically informed alarm design.

¿Y si esas y esos que andan por la vida con sus auriculares a la vista de todas y todos estuvieran haciendo algo importante pero todavía incomprendido?
Este libro sostiene que las y los audionautas son, somos, los protagonistas de una transformación muy profunda e incomprendida que, surgida en el pasaje del siglo XIX al XX, no deja de renovarse. También se afirmará aquí que, por esa incomprensión, hay dificultades para entender los alcances de la revolución digital y platafórmica que está en marcha y que nos tiene en estado de incertidumbre y expuestos a sorpresas cada vez peores. 
Estamos, en realidad, frente a la necesidad de comprender cuatro revoluciones invisibles entre las cuales, la de las audiencias en movilidad, la de los audionautas, es la resultante.
La/s radio/s, lo fonográfico, lo telefónico, diversas plataformas, nuestros parlantes y nuestros auriculares, nos traen discursos que nos rodean y en los que estamos inmersos todas y todos los que hemos nacido después de la finalización del siglo XIX.
¿Por qué revoluciones invisibles? Porque esas transformaciones revolucionarias producidas en el anterior pasaje de siglos no se destacan en nuestra cultura escritural y audiovisual. El desarrollo de las nuevas mediatizaciones, de base digital y en pantallas, parece confirmar esa tendencia de ocultamiento.

Sound and media installations are rarely considered from a time-based, formal perspective. In order to enable a greater understanding of temporal form in sound installations, I suggest a cross-disciplinary adaptation of musical form to the installation context. Due to the differences between concert and installation presentation practices – including, but not limited to, the increased agency of the mobile visitor – I re-examineformin installation contexts as the particular temporal experience co-produced by the first-person subject as they navigate in, through and out of the work’s frame. By applying this musical perspective to macro-scale formal structures, a set of tools and concepts become available for the analysis of temporal form in existing sound or audiovisual installations. Using practice-based observation and analysis, I describe several compositional strategies through which musical concepts of material and form can be extended in space and time: each of these strategies ...

A model of music listening has been automated. A program takes digital audio as input, for example from a compact disc, and outputs an explanation of the music in terms of repeated sections and the implied structure. For example, when the program constructs an analysis of John Coltrane's "Naima," it generates a description that relates to the AABA form and notices that the initial AA is omitted the second time. The algorithms are presented and results with two other input songs are also described. This work suggests that music listening is based on the detection of relationships and that relatively simple analyses can successfully recover interesting musical structure.

Computer music systems can interact with humans at different levels, including scores, phrases, notes, beats, and gestures. However, most current systems lack basic musicianship skills. As a consequence, the results of human-computer interaction are often far less musical than the interaction between human musicians. In this paper, we explore the possibility of learning some basic music performance skills from rehearsal data. In particular, we consider the piano duet scenario where two musicians expressively interact with each other. Our work extends previous automatic accompaniment systems. We have built an artificial pianist that can automatically improve its ability to sense and coordinate with a human pianist, learning from rehearsal experience. We describe different machine learning algorithms to learn musicianship for duet interaction, explore the properties of the learned models, such as dominant features, limits of validity, and minimal training size, and claim that a more h...

Human-Computer Music Performance for popular music - where musical structure is important, but where musicians often decide on the spur of the moment exactly what the musical form will be - presents many challenges to make computer systems that are flexible and adaptable to human musicians. One particular challenge is that humans easily follow scores and chord charts, adapt these to new performance plans, and understand media locations in musical terms (beats and measures), while computer music systems often use rigid and even numerical representations that are difficult to work with. We present new formalisms and representations, and a corresponding implementation, where musical material in various media is synchronized, where musicians can quickly alter the performance order by specifying (re-)arrangements of the material, and where interfaces are supported in a natural way by music notation.

This paper presents our new system Mappe per Affetti Erranti (literally Maps for Wandering Affects), enabling a novel paradigm for social active experience and dynamic molding of expressive content of a music piece. Mappe per Affetti Erranti allows multiple users to interact with the music piece at several levels. On the one hand, multiple users can physically navigate a polyphonic music piece, actively exploring it; on the other hand they can intervene on the music performance modifying and molding its expressive content in real-time through fullbody movement and gesture. An implementation of Mappe per Affetti Erranti was presented in the framework of the science exhibition "Metamorfosi del Senso", held at Casa Paganini, Genova, in October -November 2007. In that occasion Mappe per Affetti Erranti was also used for a contemporary dance performance. The research topics addressed in Mappe per Affetti Erranti are currently investigated in the new EU-ICT Project SAME (Sound and Music for Everyone, Everyday, Everywhere, Every Way, www.sameproject.eu).

Interpolated and warped digital waveguide mesh algorithms have been developed to overcome the problem caused by direction and frequency-dependence of wave travel speed in digital waveguide mesh simulations. This paper reviews the interpolation methods applicable in the two-dimensional case. The bilinear interpolation technique and two other approaches are briefly recapitulated. The use of 2-D quadratic interpolation results in a mesh structure that is efficiently implemented using only additions and binary shifts. The optimized interpolated 2-D mesh is obtained by choosing the free parameters of the interpolation method so that the difference in wave travel speed in the axial and diagonal directions is minimized. The output signal of any digital waveguide mesh can be processed with a warped FIR filter to reduce the dispersion error at all frequencies. The frequency-warping factor must be optimized individually for each mesh structure. We introduce an extended method called multiwarping, which can be used to further decrease the dispersion error. The frequency range of digital waveguide mesh simulations is also discussed.

In this work, we introduce musif, a Python package that facilitates the automatic extraction of features from symbolic music scores. The package includes the implementation of a large number of features, which have been developed by a team of experts in musicology, music theory, statistics, and computer science. Additionally, the package allows for the easy creation of custom features using commonly available Python libraries. musif is primarily geared towards processing high-quality musicological data encoded in MusicXML format, but also supports other formats commonly used in music information retrieval tasks, including MIDI, MEI, Kern, and others. We provide comprehensive documentation and tutorials to aid in the extension of the framework and to facilitate the introduction of new and inexperienced users to its usage.

Nouvelles approches des oeuvres vocales ou électroniques avec voix de Berio.
Textes de : Pierre Boulez, Olivier Class, Cyril Delécraz, Martin Feuillerac, Geoffroy Jourdain, Françoise Kubler, Philippe Lalitte, Haelim Lee, Flo Menezes, Pierre Michel, David Osmond-Smith, Valérie Philippin, Nicola Scaldaferri, Maryse Staiber, Mengqi Wang.

We present a general framework for synthesizing audio manifesting arbitrary sets of perceptually motivated, quantifiable acoustic features. Much work has been done recently on finding acoustic features that describe perceptually relevant aspects of sound. The ability to synthesize sounds defined by arbitrary feature values would allow perception researchers to more directly generate stimuli "to order," as well as providing an opportunity to directly test the perceptual relevance and characteristics of such features. The methods we describe also provide a straightforward way of approaching the problem of mapping from data to synthesis control parameters for sonification.

The process of creating an audio mosaic consists of the concatenation of segments of sound. Segments are chosen to correspond best with a description of a target sound specified by the desired features of the final mosaic. Current audio mosaicing techniques take advantage of the description of future target units in order to make more intelligent decisions when choosing individual segments. In this paper, we investigate ways to expand mosaicing techniques in order to use the mosaicing process as an interactive means of musical expression in real time. In our system, the user can interactively choose the specification of the target as well as the source signals from which the mosaic is composed. These means of control are incorporated into MoSievius, a framework intended for the rapid implementation of different interactive mosaicing techniques. Its integral means of control, the Sound Sieve, provides real-time control over the source selection process when creating an audio mosaic. We discuss a number of new real-time effects that can be achieved through use of the Sound Sieve.

The requirements of a spatial interchange format for electroacoustic music are examined with a focus on the interrelationship between the authoring of spatial content and its interpretation in audio reproduction. In context of electroacoustic music, the role of a spatial audio interchange format is to be a bridge between the processes of authoring and interpreting spatial content (Fig. ). It must enable the author to represent spatial content in a way that anticipates the means of reproduction in practical way. It must also represent the author's spatial content in a way that enables the interpreter to adapt it to a wide variety of reproduction settings. In scene-based multimedia, this interpretation is intended to be automated and to respond to user actions, but in electroacoustic music, the complexity of the interpretation is so great that one can hardly imagine how this could be accomplished except as an act of artistic judgment as happens in live diffusion. Because authoring and interpretation are so closely linked, the interchange format itself should include a description of the interpretation schemes that adapt content to the means of reproduction, even supporting alternative strategies. This is important to the author as well as the interpreter because they essentially participate in a process of artistic co-creation.

Programming has not yet become a natural medium for artists. I see two, connected causes for this: to acquire sufficient skill of computer programming and the essential algorithmic thinking is normally beyond the capacity of artists, and, for many, it is not apparent what novel possibilities the computer offers for visual exploration and for entirely new genres of experience. Processing – a recent open-source programming language made by artists for artists – makes it feasible to break both of these forbidding arguments. The goal of the workshop is to let participants – including artists who have never ever programmed before – get an insight into the programming language and its environment, practice making visual sketches and control them by using the mouse, and get an outlook on the potentials by browsing selected digital artworks. The workshop will be organized around a series of algorithmic visual tasks, to illustrate the basic algorithmic concepts and language constructs. By th...

This paper presents a new algorithmic drum sequencer, LR.step. This sequencer is based on Clarence Barlow's Indispensability algorithm, and builds upon previous work with this algorithm, introducing several novel features. LR.step differs from previous implementations of the indispensability algorithm in that it features a method for calculating arbitrary subdivisions of the beat, such as 14 th note triplets, as well as two new processes for generating syncopation. Details of the software and possibilities for future work are given.

This essay reflects on the possibility of capturing the transformation of abandoned spaces through the medium of sound. To this end, I decided to utilise Alvin Lucier’s experimental work, I Am Sitting in a Room, as both a methodology and a text on which to base my own ethnographic practice. Reciting Lucier’s words in the context of the Leith ruin, I copied his approach to (re)recording my speaking voice, so as to amplify “the natural resonant frequencies of the room” (Lucier, 1969).  Thus, I present Lucier’s work as a mode of ethnographic enquiry which reaffirms the significance of acoustic phenomenon in the rich sensorium of abandoned space. But more importantly, I suggest that sound is “good to think with” when exploring the knotty relationships between time, place, and selfhood.

This paper presents a formal model of Schoenberg's guidelines for convincing chord root progressions. This model has been implemented as part of a system that models a considerable part of Schoenberg's Theory of Harmony. This system implements Schoenberg's theory in a modular way: besides generating four-voice homophonic chord progressions, it can also be used for creating other textures that depend on harmony (e.g., polyphony). The proposed model generalises Schoenberg's guidelines in order to make them applicable for more use cases. Instead of modelling his rules directly (as constraints on scale degree intervals between chord roots), we actually model his explanation of these rules (as constraints between chord pitch class sets and roots, e.g., whether the root pitch class of some chord is an element in the pitch class set of another chord). As a result, this model can not only be used for progressions of diatonic triads, but in addition also for chords with a large number of tones, and in particular also for microtonal music beyond 12-tone equal temperament and beyond 5-limit harmony.

This paper introduces i-Berlioz, a proof-of-concept interactive Computer-Aided Orchestration (CAO) system that suggests combinations of musical instruments to produce timbres specified by the user by means of verbal descriptors. The system relies on machine learning for timbre classification and generative orchestration, and tools to write and execute interactive scenarios. The paper details the two main parts of i-Berlioz: its generative and exploratory engines, respectively, accompanied with examples. The authors discuss how i-Berlioz can aid the composition of musical form based on timbre.

This paper presents a system that uses Genetic Algorithm (GA) to evolve hierarchical pulse sets (i.e., hierarchical duration vs. amplitude matrices) for expressive music performance by machines. The performance profile for a piece of music is represented using pulse sets and the fitness (for the GA) is derived from the structure of the piece to be performed; hence the term "structural fitness". Randomly initiated pulse sets are selected and evolved using GA. The fitness value is calculated by measuring the pulse set's ability of highlighting musical structures. This measurement is based upon generative rules for expressive music performance. This is the first stage of a project, which is aimed at the design of a dynamic model for the evolution of expressive performance profiles by interacting agents in an artificial society of musicians and listeners.

Brain-Computer Music Interface (BCMI) is a new research area that is emerging at the cross roads of neurobiology, engineering sciences and music. This research involves three major challenging problems: the extraction of meaningful control information from signals emanating directly from the brain, the design of generative music techniques that respond to such information, and the training of subjects to use the system. We have implemented a proof-of-concept BCMI system that is able to use electroencephalogram information to generate music online. Ongoing research informed by a better understanding of brain activity associated with music cognition, and the development of new tools and techniques for implementing braincontrolled generative music systems offer a bright future for the development of BCMI.

The prospect of being able to gain a better understanding of how the brain processes music is very exciting for musicians and developers of music technology. Composers would certainly welcome the possibility of being able to predict more objectively the effect of particular musical configurations on their audiences. Furthermore, new music technologies are bound to emerge from such understanding. Despite an impressive amount of ongoing research into the neuroscience of music, progress in this field still remains largely uncharted for musicians and unexplored by developers of technology: the literature is complex and difficult to disentangle. This paper is an attempt to chart the field for the readership of this journal. It articulates a working hypothesis for the neural basis of mental imageries elicited by music, based on the notion that such imageries are by-products of the inherent abstracting and predicting properties of the brain. It is argued that such mental imageries are scaffolds for music perception. The paper also speculates on the impact that a better understanding of the musical brain may have on the development of future technology for electroacoustic music, which may include the development of new analysis tools such as the olivogram and the thalamogram.

As early as the 1950s and early 1960s, pioneers such as Lejaren Hiller, Gottfried Michael Koenig, Iannis Xenakis, and Pietro Grossi, among a few others, started to gain access to computers to make music. It soon became clear that to render music with a socalled "human feel," computers needed to process information about performance (e.g., deviations in tempo and loudness), in addition to the symbols that are normally found in a traditional musical score (e.g., pitch and rhythm). This was especially relevant for those interested in using the computer to play back scores. Indeed, the first ever attempt at creating a computer-music programming language, by Max Mathews at Bell Telephone Laboratories in 1957, was motivated by his wish to "write a program to perform music on the computer" (Park 2009 p. 10). It appears that this development began after Mathews and John Pierce went to a piano concert together. During the intermission, Pierce suggested that perhaps a computer could perform as well as the pianist. Mathews took up the challenge, which resulted in Music I, the ancestor of music programming languages such as Csound . Research into computational models of expressive performance of music (Widmer and Goebl 2004) is still an active area of study-particularly, research into devising increasingly more sophisticated automated and semi-automated computer systems for expressive music performance, hereinafter referred to as CSEMP. A CSEMP is able to generate expressive performances of music. For example, software for music typesetting is often used to write a piece of music, but most packages play back the music in a

We are interested in exploring ways in which unconventional modes of computation may provide new directions for future developments in Computer Music. Research into Unconventional Computing (Calude et al. 1998) is aimed at computational paradigms other than the standard von Neumann architecture, which have prevailed in computing since the 1940s. This paper presents an investigation into the feasibility of synthesizing sounds with hybrid wetware-silicon devices using in vitro neuronal networks. The field of Computer Music has evolved in tandem with the field of Computer Science.

We propose a model of expressive music performance (EMP), focusing on the emergence of EMP under social pressure, including social interaction and generational inheritance. Previously, we have reported a system to evolve EMP using Genetic Algorithm, exploring the effect of generational inheritance. This paper presents a system that evolves expressive performance profiles through social interaction, with a society of artificial agent performers. Each performer owns a hierarchical pulse set (i.e., hierarchical duration vs. amplitude matrices), representing a performance profile for a given piece. An agent performer evaluates a performance profile with a set of rules derived from the structure of the piece in question, and imitates others' performances if appropriate. Then it modifies its pulse set accordingly. We demonstrate that suitable performance profiles emerge from social interactions where the diversity and the commonality of evolved performances are observed in the society of agents.

This paper introduces a system that uses brainwaves, or EEG (electroencephalogram), information to compose and play music in real-time. The system composes music using generative grammars and transition rules controlled by means of information extracted from the EEG of the subject. The paper starts by noting the various attempts at the design of systems that make music from the EEG signals, followed by a short technical introduction to the EEG. The generative component of the system is inspired by the work of David Cope [7] on computer-replication of a musical style by analysis of given musical examples. The paper concludes with a brief discussion on the contribution of our research for the development of assistive technology for physical disability.

We present a survey of research into automated and semiautomated computer systems for expressive performance of music. We will examine the motivation for such systems and then examine the majority of the systems developed over the last 25 years. To highlight some of the possible future directions for new research, the review uses primary terms of reference based on four elements: testing status, expressive representation, polyphonic ability, and performance creativity.