installation during my residency @massmoca summer 2016.
nn shows a physical system that is an embodiment of a neural network in both sound and form. The sounds were rendered using recurrent neural networks (RNN) to synthesize sounds. In this type of machine learning, recursive neural nets literally teach the computer how to produce sounds representative of the machines themselves. As such, this piece focuses on a corpus of field-recorded sounds such as industrial drones, refrigerator hums, consumer electronics noise, computer fan noise, etc. After processing these recordings, the computer ‘dreams up’ sound based off of its own idea of what industrial noise is and in effect, can be thought of attempting to listen to itself. In this sonic assemblage of noise-classified audio, the output sound is a mixture of both real-life analog noise and the computer’s interpretation of the same. The sounds undulate, swell, and breathe to form an ecology of machine-interpreted awareness, one that suggests a strange convergence of the real and the digitally imagined, the sentient and the synthetic.
The physical installation sees an array of light switches that are tethered to strings, balanced by small rocks. Light switches are an industrial found object and physical signifiers of a binary process, off or on. In this context, their functionality is subverted to reveal their movement as continuous, in between states, artificially intelligent in their communicative gestures. As the natural light changes in the space, the movement of the light switches suggests an inversion of mechanistic process: the light switches now control the atmospheric light. As such, each switch can be thought of as a node within the network, one that works to balance a complementary set of weights.