Intuitive Composing in Jazz by Sara Simionato

AI & Music Jazz Composition Jazz Improvisation Somax2

The research project Intuitive Composing in Jazz explores the integration of AI technologies into the creative process of jazz improvisation and composition, utilizing AI-tools such as RAVE and Somax2 as co-creative agents. The resulting musical output is recorded, transcribed and analyzed to investigate how intuition and embodiment can be affected in the human-machine interactive co-creative process, and ultimately, to generate new music scores. While AI has been explored in various musical genres, including classical, electronic, and popular music, its application to jazz remains generally unexplored in an artistic research context. Although the jazz compositional process often employs improvisation in its intuitive and embodied nature, there is a lack of literature focusing on these aspects; furthermore, there is a gap in knowledge regarding embodiment and creativity in musical AI. Building on the ongoing research of the CREATIE research group, the project aims to improve and develop new approaches, methodologies and skills regarding the human-machine co-creative process in jazz improvisation and composition.

➡️ This presentation is part of IRCAM Forum Workshops Paris / Enghien-les-Bains March 2026

This research project explores the integration of AI technologies into the creative process of jazz improvisation and composition, utilizing Somax2 as a co-creative agent. The central research question is: How do AI-driven systems influence intuitive and embodied creativity in jazz improvisation and composition?

While AI has been explored in various musical genres, including classical, electronic, and popular music, its application to jazz remains generally unexplored in an artistic research context. Although the jazz compositional process often employs improvisation in its intuitive and embodied nature, there is a lack of literature focusing on these aspects; furthermore, there is a gap in knowledge regarding embodiment and creativity in musical AI. This project addresses these gaps, exploring the human-machine relationship in an embodied creative context, and suggesting new possibilities for jazz improvisation and composition.

The methodological approach involves improvised music sessions that are recorded, transcribed and analysed to investigate how intuition and embodiment can be affected in the human-machine interactive co-creative process, and ultimately, to generate new music scores. In this framework, artistic practice serves as the primary method of investigation: the process of improvising with AI tools offers experiential, analytical and reflective knowledge, while the transcription and analysis of improvised music sessions offer a way to question, test and refine the methodologies throughout the research. In addition, semi-structured interviews are conducted with musicians and experts in musical AI interaction. These interviews provide further insight into performers’ embodied experience of human–machine co-creation and inspire new understandings and methodologies inside the process.

Building on the ongoing research of the CREATIE research group, the project aims to improve and develop new approaches, methodologies and skills regarding the human-machine co-creative process in jazz improvisation and composition.