François Pachet (born 10 January 1964 in Sainte-Adresse, France) is a French scientist, composer, and director of the Spotify Creator Technology Research Lab, located at 166 rue du faubourg Saint-Honoré, 75008 Paris. Associated with Pierre and Marie Curie University, he previously led the Sony Computer Science Laboratory in Paris before joining Spotify.
He is one of the pioneers of computer music closely linked to artificial intelligence, especially in the field of machine improvisation and style modelling, known for projects including Continuator and Flow Machines. He was elected an ECCAI Fellow in 2014 and received a best paper award in 2002 for "The Continuator: Musical Interaction with Style".[1] In 2025, he cofounded https://imagineallthepeople.io, a pioneering company dedicated to synthetic populations, as well as Ynosound, a startup dedicated to AI-assisted songwriting and composition.
Education
Pachet graduated from École des ponts ParisTech in civil engineering and computer science in 1987, majoring in applied mathematics. In 1987–1988, he spent 18 months as a lecturer at the University of Malaya in Kuala Lumpur. He obtained a PhD in computer science from Pierre and Marie Curie University, with the thesis "Knowledge representation with objects and rules: the NéOpus system", supervised by Jean-François Perrot. He subsequently spent one year as a post-doctoral researcher in Montréal at Université du Québec à Montréal, where he worked on Douglas Lenat's Cyc project on common sense representation with the assistance of UQAM professor Hafedh Mili.[2,3] In 1997, he obtained his habilitation diploma on the subject "Object-oriented languages and knowledge representation" at Pierre and Marie Curie University.[2,3] In 2006, he was an auditeur at the 58th national session of the Institut des Hautes Études en Défense Nationale, and in 2007, he was appointed Colonel in the "réserve citoyenne" of the French Air Force.[2,3]
Experiences
In 1993, he was appointed Assistant Professor (in French, "Maitre de conférences") in Computer Science, Research and Teaching at Pierre and Marie Curie University, where he remained until 1997. Pachet moved to Sony-CSL (Computer Science Laboratory) Paris in 1997, where he started a research activity on music and artificial intelligence. The CSL, a branch of Sony-CSL Tokyo dedicated to basic research in computer science, was created by Luc Steels and Mario Tokoro in 1996. His team has authored and pioneered many technologies (about 35 patents) related to electronic music distribution, audio feature extraction, and music interaction. He was appointed director of Sony Computer Science Laboratories in 2014.
Since 2017, he is director of Spotify's Creator Research Technology Lab in Paris, where he develops tools for assisting music creation.[4]
Achievements
The Music team at Sony Computer Science Laboratory Paris was founded in 1997 by Pachet, where he developed the vision that metadata can greatly enhance the musical experience, from listening to performance.[5] The "Popular Music Browser" project, started in 1998 at Sony Computer Science Laboratories, covered all areas of the music-to-listener chain, from music description, descriptor extraction from the music signal, or data mining techniques, to similarity-based access, novel music retrieval methods such as automatic sequence generation, and user interface issues.[7] In 2000, he created MusicSpace, a spatialization control system, with O. Delerue. Another achievement is CUIDADO (Content-based Unified Interfaces and Descriptors for Audio/music Databases available Online), a two-year project ending in 2003 on developing content-based audio modules and applications, including analysis, navigation, and the creative process.[13,14] This project satisfied the needs of record labels and copyright societies for information management methods, marketing, and protecting their information, utilizing an authoring system using content features for professional musicians and studios.
His most notable achievement is the Continuator, an interactive music improvisation system designed to allow real-time musical improvisation with an algorithm.[9,8,16,15] Experimented with by many professional musicians and presented notably at the SIGGRAPH'03 conference, the Continuator is considered a reference in the domain of music interaction, including an example of a Musical Turing test conducted on the VPRO Channel with Amsterdam jazz pianist Albert van Veenendaal.[17] The Flow Composer is his second achievement, designed as a system to compose lead sheets in the style of arbitrary composers. It was followed by LSDB, the first large-scale effort collecting lead sheets in electronic format with over 11,000 lead sheets collected, as well as Virtuoso, a solo jazz detector. He became the beneficiary of the ERC Grant Flow Machines for investigating how machines can boost creativity in humans and be able to continue a work in the same musical style.[10] Envisioning a future where consumers could buy the unique style of an artist and apply it to their own material, Pachet stated, "I call it 'Stylistic Cryogenics' -- to freeze the style into an object that can be reused and made alive again".[11]
In 2014, Pachet presented two music tutorials on Brazilian guitar and Jazz. In September 2015, he wrote about using CP techniques to model style in music and text for the Association for Constraints Programming (ACP). ARTE presented Pachet on Square Idée in "Demain, devenir Wagner ou Daft Punk?" ("Tomorrow, become Wagner or Daft Punk?") in October 2015.[18] In 2017, he produced and released Hello World, a multi-artist album composed with Artificial Intelligence.[19]
Bibliography
F. Pachet received a best paper award for 'Playing with Virtual Musicians: the Continuator in practice', published in IEEE MultiMedia in 2002. Pachet's subsequent publications include 'The future of content is in ourselves' in Computers in Entertainment in 2008, 'Description-Based Design of Melodies' in Computer Music Journal in December 2009, and 'Markov constraints: steerable generation of Markov sequences' in Constraints in March 2011. In 2012, Pachet published 'Computers and Creativity' with Springer, with additional publications linked to Sony CSL.
Personal Writings
Pachet has written several non-scientific books about music. These include Histoire d'une oreille, an augmented book about the ontogenesis of a musical ear, and Comment je n'ai pas rencontré Paul McCartney, a short story. He also authored Max Order ou l'invention du style, a comic book realized in the context of the ERCcOMICS project.
External links
External links include François Pachet's profile at Sony CSL Paris, François Pachet's personal website, and the Flow Machines website. Additionally, external media includes the 2015 Hot Topics feature, "Can being a jazz musician make you a better decision maker?".
