Creative machine learning in MaxMSP: Concepts, tools and applications
Objectives
À l'issue de ce cours, les stagiaires seront capables de comprendre les bases du fonctionnement des IA génératives, d'utiliser différents modèles audios existants, et de personnaliser par la suite les modèles à leur applications créatives. La formation se termine par un travail personnel et une restitution permettant aux participants de développer l’usage de ses outils dans leur projet créatif.
Public
Composers, musicians, teachers, performers, researchers.
Prerequisites
- Fluent in spoken and written English.
- Les participantes et participants doivent être à l'aise avec les environnements de MAO (MaxMSP, Ableton Live) et avoir une expérience dans la composition avec ces outils.
- Basic knowledge of Max and its environment: hands-on familiarity and understanding of Max’s basic tutorials (Max Basics tutorials 1–11, 14–22; Max Data tutorials 1 and 2; MSP tutorials; MSP Basics tutorials 1–5; MSP Panning tutorials 1–3; MSP Analysis tutorials 1 and 2; MSP Delay tutorials 1–3 and 6), or completion of the Max Beginner course
- A beginner’s level knowledge of Python is strongly recommended.
- Knowledge of signal processing basics is a plus.
- Experience in mathematics and computer programming is a plus.
- Basic concepts of using the Zoom application.
Training Period and Organization
8 hours of training. Every Tuesday from March 3 to March 24, 2026 , from 2 pm to 4 pm.
In English, Online Class
8 students maximum
Program
Some details of the training program are subject to change depending on developments in the software on which they are based.
After a brief introduction to the concepts of artificial intelligence and machine learning, participants will delve into the mechanisms and practical use of nn~ and RAVE, then learn how to customize them for their artistic projects.
Tuesday, March 3
Core curriculum Part 1:
- Fundamentals of machine learning and generative AI for music.
- Introduction to RAVE VST and nn~ in MaxMSP and Ableton.
Tuesday, March 10
Core curriculum Part 2:
- Theoretical deep dive: latent space and transformations.
- Creation of control patches for different models in MaxMSP.
Tuesday, March 17
This session will be divided into two groups to allow for more personalized learning:
Group 1: Training your own generative models
- Getting started with Pytorch and RAVE training code in Python.
- Scripting pre-existing models for integration into nn~.
Group 2: Advanced control and bending
- Creating complex patches for latent manipulation.
- Motion control.
- Network bending.
At the end of the session, time will be set aside for trainees to begin work on a personal application, in order to put the concepts studied into practice, with a view to presenting their work the following week.
Tuesday, March 24
Presentation of personal work, discussion, and ideas for further exploration.
Educational Resources and Techniques
- This professional training course will be held online and taught in English.
- Technical requirements: Please download the following software prior to the training course:
- To get the most out of the training, you will need two screens and a good internet connection.
- Links to access the platform and instructions for taking the course remotely will be provided to each participant.
- The software is not included in the training fee and will not be provided at the end of the training.
Supervision
- Attendance controlled : signature of an attendance sheet required every morning and afternoon.
- A certificate at the end of the program is given to each participant with the results of the exam.
.Assessment
- Assessment of knowledge and skills in the form of an open creative project.