A grand piano, a computer, two microphones. And a loop.

ImprovIA is what happens when one of the most relevant jazz pianists in Spain begins to improvise and an artificial intelligence system listens to him with the same attention any other musician on stage would. The machine extracts acoustic and musical parameters from what it hears, generates its own improvisation, and returns it to the room. Then Terraza listens and responds. The machine listens again and responds. And so on, until it is no longer clear who proposed what, or in what order.

The piece that emerges respects no style. It moves through jazz, through extended piano, and through territories that do not yet have a name, because the loop itself is opening them up. ImprovIA does not use AI as a generator of sonic backgrounds or as a producer of effects: it uses it as a co improviser.

Ignasi Terraza is one of the most prominent jazz pianists in Spain, with a personal style that fuses the jazz tradition with classical influences. An extensive discography, collaborations with leading international figures, and an uncommon combination of virtuosity, swing and great expressive sensitivity.

Philippe Salembier is a professor at the Department of Signal Theory and Communications at the Polytechnic University of Catalonia (UPC), with a long trajectory in signal and image processing. He directed the Piano & AI project with Marco Mezquida, presented at the S+T+ARTS AI and Music Festival, and is currently expanding his AI system in collaboration with Terraza.

Two ways of listening on the same stage, and an audience in the middle asking one question: who is improvising with whom?