human-machine co-exploration

This project consisted in studying how people partner with artificial agents in parameter space exploration for sound design.

Artificial Intelligence techniques (AI) have powerful computational abilities that could help human interaction with the digital world. Yet, care should be taken on how to combine AI with human intelligence to not simply replace humans. This may be crucial in creative contexts where full automation may not always benefit the emergence of ideas.

We led a workshop to investigate how professional creatives may partner with AI, using our « co-explorer » software to explore sound synthesis spaces. We observed various degrees of appropriation among users, sharing the lead with AI at different stages of exploration to foster their creative process.

human-computer interaction
machine learning


The project was developed with Bavo Van Kerrebroeck, Baptiste Caramiaux, and Frédéric Bevilacqua in collaboration with the ISMM group of IRCAM and the ex)situ team of LRI (INRIA), in the context of the Sorbonne Université Doctorate in Computing.

arXiv preprint (2019)


Ce diaporama nécessite JavaScript.