The Machine has made itself, quietly, unnoticed, in a back room. From partially broken and discarded components. Now it tries to explore and understand the world in which it finds itself. It tastes what it sees - abandoned pieces of junk, rotting matter. In response, tentatively, it begins to express itself, in machinic love poetry and soundscape.
There is a surprisingly long history of machines composing their own poetry. Early in the nineteenth century Ada Lovelace had the insight that the numbers manipulated by a computer could stand as symbols for anything, including music or words. This opens up not only the possibility of using a computer creatively, but also for synaesthesia: the symbols can shift their expression from one domain to another - from visual image, to taste, to words, to sound.
Christopher Strachey, who worked alongside Alan Turing on the Manchester University Computer in the early 1950s, programmed the computer to create love letters, which he fixed to the lab walls. This Machine follows in that tradition, using machine learning. A neural network has been trained to associate tastes (salty, bitter, sweet, sour, umami) with a video input stream. Those perceived tastes trigger the choice of words from ‘flavoured’ pools of vocabulary taken from Shakespeare’s sonnets and WIRED online; the lines of poetry evolve gradually towards sonnet structures.
The Machine’s software was written in C++ and openFrameworks. It passes each video frame to a neural network in Wekinator to taste the video stream. The tastes are passed to a genetic algorithm which generates the soundscape and fragments of poetry, scoring each by analysing sentences using the Stanford University Natural Language Group’s part-of-speech tagger. Two Arduino boards control a thermal printer, video camera and motor.
The audience are invited to interact with the installation to influence outcomes, to feed the Machine: add their own bits and pieces, an old apple core, the sweet wrapper or fluff at the bottom of a pocket; mix it up; and perhaps take away some printed fragments of poetry.
Laura Dekker is a British artist whose work explores the reciprocal roles of technologies in how we experience, make sense of, and construct ourselves and our world.
Her interactive installations combine physical materials, layered video, audio, robotics and machine learning. She aims to engage the viewer-participant with a sensorially rich and provocative experience: virtual objects can intrude into the ‘actual’ world, or objects are activated with a kind of primitive consciousness. There is always a performative aspect - at the point of production, reception, or both.
Before becoming an artist, Laura worked for many years as a research scientist in 3D imaging and artificial intelligence.
Her work is shown internationally at festivals, museums, parkland, historical sites and urban public spaces. She was selected for the Lumen Prize world tour (2014-6), and invited to join pioneering digital women artists in 'Technology Is Not Neutral' (2016) and 'V&A Digital Futures' (2017).
Laura often works collaboratively with the London-based collective, XAP, a group of artists who combine performance, traditional art media, video and digital installation.