Search

Computer-based generative approaches possess a great creative potential in Contemporary Dance, in particular for artistic realisations that combine dance and technology. At the same time, the adoption and dissemination of generative approaches in dance is hampered by the fact that Dance and Technology is a small subfield within Contemporary Dance, with Generative Dance occupying an even smaller niche within this subfield. The work presented in this paper tries to ameliorate this situation by supporting artistic communities in Contemporary Dance and Generative Art with practical resources in the form of source code, dance data, educational articles, and documentations of exemplary artistic realisations. This material is meant to motivate and facilitate the selection from and adoption of a wide range of computational techniques and their use as foundations for realising dance specific generative systems. These techniques include both computer simulations and machine learning models that have proven useful in the author's own collaborations with dancers and choreographers for translating embodied creation principles into generative procedures. With regards to the integration of generative systems into the creative process, the provided material differentiates itself from other existing tools and collections in that it supports artists in devising their own idiosyncratic generative systems instead of working with a readily available but inscrutable software., + ID: 591463

This paper describes the authors' first experiments in creating an artificial dancer whose movements are generated through a combination of algorithmic and interactive techniques with machine learning. This approach is inspired by the time honoured practice of puppeteering. In puppeteering, an articulated but inanimate object seemingly comes to live through the combined effects of a human controlling select limbs of a puppet while the rest of the puppet's body moves according to gravity and mechanics. In the approach described here, the puppet is a machine-learning-based artificial character that has been trained on motion capture recordings of a human dancer. A single limb of this character is controlled either manually or algorithmically while the machine-learning system takes over the role of physics in controlling the remainder of the character's body. But rather than imitating physics, the machine-learning system generates body movements that are reminiscent of the particular style and technique of the dancer who was originally recorded for acquiring training data. More specifically, the machine-learning system operates by searching for body movements that are not only similar to the training material but that it also considers compatible with the externally controlled limb. As a result, the character playing the role of a puppet is no longer passively responding to the puppeteer but makes movement decisions on its own., + ID: 591443

Raw Music from Free Movements is a deep learning architecture that translates pose sequences into audio waveforms. The architecture combines a sequence-to-sequence model generating audio encodings and an adversarial autoencoder that generates raw audio from audio encodings. Experiments have been conducted with two datasets: a dancer improvising freely to a given music, and music created through simple movement sonification. The paper presents preliminary results. These will hopefully lead closer towards a model which can learn from the creative decisions a dancer makes when translating music into movement and then follow these decisions reversely for the purpose of generating music from movement., + ID: 591439 + PeerReviewed

Raw Music from Free Movements is a deep learning architecture that translates pose sequences into audio waveforms. The architecture combines a sequence-to-sequence model generating audio encodings and an adversarial autoencoder that generates raw audio from audio encodings. Experiments have been conducted with two datasets: a dancer improvising freely to a given music, and music created through simple movement sonification. The paper presents preliminary results. These will hopefully lead closer towards a model which can learn from the creative decisions a dancer makes when translating music into movement and then follow these decisions reversely for the purpose of generating music from movement.

Strings is an audiovisual performance for an acoustic violin and two generative instruments, one for creating synthetic sounds and one for creating synthetic imagery. The three instruments are related to each other conceptually, technically, and aesthetically by sharing the same physical principle, that of a vibrating string. This submission continues the work the authors have previously published at xCoAx 2020. The current submission briefly summarizes the previous publication and then describes the changes that have been made to Strings. The P in the title emphasizes, that most of these changes have been informed by experiences collected during rehearsals (in German Proben). These changes have helped Strings to progress from a predominantly technical framework to a work that is ready for performance.

The project Klavierspiel ('piano playing') combines a piano automaton and interactive, touch-screen-based control software. The piano automaton is a piano-playing robot to be fixed on the keyboard of any conventional instrument. It can perform much faster movements and strike many more keys at the same time than a human pianist. Therefore, it lends itself to a variety of sonic experiments, especially for computer-generated music. The authors developed three different graphical user interfaces to control the automaton. Explicitly aimed at people without musical expertise, these interfaces provide the opportunity to gain hands-on experience with generative music. They illustrate how to create musical gestures, patterns and structures at different levels of abstraction, and convey specific algorithmic composition techniques in an easy-to-understand and practice-oriented way. This paper describes the three user interfaces regarding their technical implementation and their musical potential. Furthermore, it discusses the observations made during a three-day exhibition in Zurich., + ID: 571651

This publication reports on preliminary experiments in employing smart materials as central elements in a generative sound installation. For this installation, a material has been chosen whose piezoelectric properties can be altered by changing the material's shape. When integrated into a feedback loop, the material's capability to simultaneously act as an acoustic sensor and actuator establishes a sustained audio signal whose timbre is dependent on the bending of the material. By exploiting this material behaviour, a process can be set into motion within which the material serves both as tangible interface for interaction and as generative mechanism that drives the development of the musical output. This work continues a research trajectory that explores the generative potential of physical processes and their integration into hybrid artworks in which physical and computational systems complement each other. Throughout the course of this research, the emphasis has increasingly been placed on the characteristics of physical processes while relegating the role of computation to one of control and stabilization. This installation demonstrates the potential usefulness of this approach but also highlights its pitfalls, the most important of which is the difficulty of controlling and scaling the complexity of a physical process through an incremental development method., + ID: 572473

PIANO & DANCER is an interactive piece for a dancer and an electromechanical acoustic piano. The piece presents the dancer and the piano as two performers on stage whose bodily movements are mutually interdependent. This interdependence reveals a close relationship between physical and musical gestures. Accordingly, the realization of the piece is characterised by a creative process that merges choreographic and compositional methods. In order to relate the expressive movement qualities of a dancer to the creation of musical material, the piece employs a variety of techniques. These techniques include methods for movement tracking and analysis, generative algorithms for creating spatial and temporal structures, and the application of non-conventional scales and chord transformations to shape the modal characteristics of the music. The publication presents the artistic and technical aspects of this work and discusses some of the challenges that have shaped the creative outcome.

Perception of gesturality in music performance is a multi modal phenomenon and is carried by the differentiation of salient features in movement as well as sound. In a mix of quantitative and qualitative methods we collect sound and motion data Laban effort qualifiers and in a survey with selected participants subjective ratings and categorisations. The analysis aims at uncovering correspondences in the multi modal information using comparative processes to find similarity/differences in movement sound as well categorical data. The resulting insights aim primarily at developing tools for automated gestural analysis that can be used both for musical research and to control interactive systems in live performance.

This article explores modes of interaction or ‘interfacing’ between dancers or musicians and algorithms and the ways in which inter dependence and co performance between human and machine performers arises. Two artistic works an interactive dance piece and a live electronic music piece serve as example cases for the observation and discussion of how algorithms can play a central role in composition processes as well as human machine interactions during performance. The intended goal of using algorithms in this context is the emergence of idiosyncratic behaviours in interactive systems that reflect and combine aspects of the performance situations the generative algorithms and the adaptation mechanisms themselves. By comparing the interaction relationships and ‘interfaces’ of the example works fundamental differences of the algorithmic systems become visible and a better understanding of the impact and effect of algorithmic systems in real life performance can be gained.

When performing with gestural devices in combination with machine learning techniques, a mode of high-level interaction can be achieved. The methods of machine learning and pattern recognition can be re-appropriated to serve as a generative principle. The goal is not classification but reaction from the system in an interactive and autonomous manner. This investigation looks at how machine learning algorithms fit generative purposes and what independent behaviours they enable. To this end we describe artistic and technical developments made to leverage existing machine learning algorithms as generative devices and discuss their relevance to the field of gestural interaction.

This paper presents a proof of concept implementation of an interface entitledCoral. The interface serves as a physical and haptic extension of a simulatedcomplex system, which will be employed as an intermediate mechanism for thecreation of generative music and imagery. The paper discusses the motivationand conceptual context that underly the implementation, describes its technicalrealisation and presents some first interaction experiments. The paper focuseson the following two aspects: the interrelation between the physical andvirtual behaviours and properties of the interface and simulation, and thecapability of the interface to enable an intuitive and tangible exploration ofthis hybrid dynamical system.

In this paper an audio-visual installation is discussed, which combines interactive, immersive and generative elements. After introducing some of the challenges in the field of Generative Art and placing the work within its research context, conceptual reflections are made about the spatial, behavioural, perceptual and social issues that are raised within the entire installation. A discussion about the artistic content follows, focussing on the scenography and on working with flocking algorithms in general, before addressing three specific pieces realised for the exhibition. Next the technical implementation for both hardand software are detailed before the idea of a hybrid ecosystem gets discussed and further developments outlined.