PhD Candidate – AI-Assisted Musical Composition

Full-time, 36-month fixed-term doctoral contract – PostGenAI Project. Start date : Early 2027

IRCAM is a research and creation institute associated with the Centre Pompidou (Centre national d’art et de culture Georges Pompidou), whose activities combine scientific research, technological development, musical creation and education. Within the Sciences and Technologies for Music and Sound (STMS) laboratory, hosted at IRCAM, specialized teams conduct research in acoustics, sound and music signal processing, machine learning, interaction, computer music and musicology.

In the context of the PostGenAI project, led by Sorbonne University and operated through its Sorbonne Cluster for Artificial Intelligence (SCAI), IRCAM is recruiting a PhD candidate to conduct a doctoral thesis on audio generative models for AI-assisted music composition. The successful candidate will work in collaboration with researchers in machine learning, audio signal processing and computer music, as well as composers and music practitioners, and is expected to enroll as a PhD student at Sorbonne’s Ecole Doctorale Informatique, Télécommunications et Électronique (EDITE).

The position will be hosted by the Analysis/Synthesis team, which conducts research on advanced signal processing and machine-learning methods for speech and music.

The PostGenAI project :

PostGenAI is an interdisciplinary AI project led by Sorbonne University through the Sorbonne Cluster for Artificial Intelligence (SCAI). Supported by the French government under the France 2030 program, it gathers academic institutions, public organizations and industrial partners with the ambition of establishing an international hub of excellence dedicated to post-generative AI. This new era of AI moves beyond mere content generation to foster deeper understanding and more controlled interactions with its environment.

Role of IRCAM in the project :

IRCAM contributes to PostGenAI as leader of the AI-MADE (AI for Music And voice meDia gEneration) program. It aims to develop innovative generative models for professional audio content generation and manipulation. The central scientific problem to be investigated is the intuitive and precise control of sound properties, notably related to translating from idea/concept into sound. Within the project, the Analysis/Synthesis team will investigate generative models designed to support professional music composition, with an emphasis on fine-grained and multimodal control, adaptation to composers' own musical material, and computationally accessible models.

Composition is an iterative process in which musical ideas are explored, transformed, extended and revised. The objective of the PhD is therefore to develop audio generative models as controllable tools for music composition and sound design, supporting such operations while preserving fine-grained creative agency. As musical intentions can be expressed through various modalities (audio, scores, text, etc.), the PhD will also investigate generative models capable of combining such complementary forms and scales of conditioning. Finally, as generative models require sufficiently large corpora to learn a rich musical prior, while composers need models that can adapt to their own material and artistic language, the PhD will also investigate efficient training and fine-tuning strategies for controllable, composer-specific music generation.

Master's degree or equivalent covering machine learning & audio signal processing.
Strong programming skills in Python and pytorch, good knowledge of GNU/Linux.
Experience in music information retrieval and computer-assisted music composition is a plus.
High productivity, ability to conduct methodical and autonomous research, develop high-quality research software, communicate scientific results clearly, and collaborate with researchers and musical practitioners.

Gross monthly salary : €2,400
32 days of paid annual leave and 11 days RTT, based on full-time employment for a complete year.
Meal vouchers (titres-restaurant): €11 per working day.
Public transport : 50% reimbursement of public transportation costs.
Health and insurance benefits: Comprehensive group health insurance and attractive employee benefits coverage.
CSE : Access to cultural, sports and social activities.
Culture card: Free access to numerous cultural venues across France for you and one accompanying person.

Applications should include a CV, a motivation letter, and any suitable information addressing the above elements preferably by email to: ai_assisted_composition_phd@ircam.fr

Candidature

Applications open until 15 December 2026

Job offer details

ContractFixed-term contract

DatesDate to be confirmed

DurationFull-time

Location75004 Paris

Our Commitments

At IRCAM, we believe that artistic and scientific innovation can only grow in an environment that is respectful, inclusive, and sustainable. Our CSR (Corporate Social Responsibility) policy shapes all our actions.

Our Commitments

PhD Candidate – AI-Assisted Musical Composition

About the role