My PhD thesis introduced the notions of machine expression and co-expression to describe the musical interactions created by human-centred machine learning.
The notion of expression is not new in the field of Computer Music. Some researchers even focused on expressiveness as an essential feature of interactive music systems, notably through the spreading of NIMEs—acronym for New Interfaces for Musical Expression. Yet, we believe that the « E » in NIME has mainly been considered for members of its computer music community—e.g., expressiveness for expert performers, or for expert audience members. We are interested in including more people in the notion of musical expression, especially through the design of interactive music systems.
The notion of machine expression aims at describing interaction with interactive music systems from a human-centred point of view—be they musicians, or non-musicians. Rooted in embodied music cognition, machine expression pragmatically addresses the fact that humans may perceive expression in machines, regardless of machines’ abilities to express by themselves. As such, machine expression may help design for a wider range of human creative processes in relation to various musical activities, such as motion-sound mapping, sonic exploration, synthesis exploration, or collective musical interaction.
PhD thesis (2020)