Crosstalk: Making people in interactive spaces

Crosstalk is an interactive performance work by media artist Simon Biggs, choreographer Sue Hawksley and composer Garth Paine. The work employs real-time multi-modal sensing and interaction systems, including three-dimensional tracking of multiple performers combined with multi-source voice recognition for speech to text and an interactive multi-channel data driven sound score. The three artists have previously collaborated on Bodytext, an interactive multimedia solo performance work in which a spoken (described) and performed dance are simultaneously interpreted by both the performer and the computational system. Crosstalk developed out of the processes undertaken in Bodytext, specifically the 'drama of the performance', which arose from an antagonistic but interdependent human/machine relationship. Created for two performers, Crosstalk engages social relations as articulated through performative language acts. The project explores ontologies of self-hood within the generative potential of a technologically mediated social space. The elements in the system, including performers and machines, affect how each adapts from state to state, as the various elements of the work - language, image, movement and sound - interact with one another. Developed as an enactment of the proposition of 'making people', inspired in part by contemporary anthropological ideas, Crosstalk begins with two dancers speaking descriptions of each other. Automatically transcribed in real time into a virtual threedimensional world, using speech to text software, these descriptions become textual objects that inhabit the environment and interact with other elements within the system, both human and non-human. The spoken texts form the foundation for an evolving sonic environment. When moving the performers collide with the text objects, causing them to also move. As the text objects interact, they re-write each other, facilitating the emergence of new textual and sonic material, created through the recombinant computation of the texts in the collided objects. Through this generative mechanism, the interaction situates each dancer as a product of their initial perception, the evolving environment, their interaction with it and their interrelationships. Crosstalk thus presents the multidimensional emergent properties of perception, interaction, place making and identity. In both Crosstalk and Bodytext, the performers are enmeshed in a public/private drama within an interactive system, to which the audience is witness. Movement gesture is further enacted through multi-channel spatialisation of audio, enveloping the audience in the 'dance as sound' through the placement of the loudspeakers around them. As the dancers move, bodily and vocal sounds are acquired and processed, the resulting sounds and texts dynamically shifting around the audience. The spatialisation of texts and sounds immerse the audience within the morphology of the dancers' gesture and the act of revealing the dancers' inner voice. In some presentation scenarios (such as art gallery settings) the audience members can interact directly with the Crosstalk system and become part of the ecology of the work.