The Sound of Speech As It Echoes in the Cloud

Dive into a more-than-human sonic assemblage

Faced with the parallel tensions of ecological emergency and global pandemic, it is sensible to hope for new ways of co-inhabiting our damaged Earth ecosystem. As most of our existing models have failed to ensure fertile tomorrows, we advocate for divesting ourselves of our too-human systems and developing new modes of interacting with the more-than-human worlds. Within this context, The Sound of Speech As It Echoes in the Cloud confronts us with a semantic otherness of multiple non-human agencies through voice technologies. The project presents a voice assistant network with geographically distributed nodes; each node is embodied in a physical form, as objects dwelling in domestic and public spaces. Humans are invited to join this dynamic assemblage, which will, in turn, learn, fabulate, and evolve from these exchanges. Building on the open-source framework Mycroft and generative machine learning models such as GPT-2, the voices capriciously scrape, generate and share ecological speculations and inquisitions, tangible practices of making kin, or some bursts of collective memory across spacetime. This project intends to bring echoes and frictions of trans-border and trans-species collectives into our lives, and open up new aesthetic and affective dimensions.