Github profile:


Modular Interactive Machine Learning Toolkit

Marcelle is a set of modules written in javascript for Interactive Machine Learning. The toolkit allows user to build reactive interfaces for ML pipelines. Pipelines can be written in javascript (we build a series of modules based on tensorflowjs) or in python. The toolkit has been built by Jules Françoise, Téo Sanchez and myself

Marcelle website:


Cortical Timbre Library

The coritcial timbre library is a set of tools developed mostly in Python used in the article "Learning metrics on spectrotemporal modulations reveals the perception of musical instrument timbre" published in Nature Human Behaviour (link). The software has been developed in collaboration with Etienne Thoret.

Github repo: (python, matlab)


Gesture Variation Follower (GVF)

GVF is a C++ library for realtime gesture recognition and realtime tracking of gesture variations. The library can be interfaced in creative programming environment such as Max/MSP, PureData or openFrameworks. Examples are provided online. The GVF library is under LGPLv3 licence.

Github repo: (C++, MaxMSP, PureData)
See also the academic paper


Gestural Sound Toolkit

The toolkit gathers a collection of Max/MSP patches and objects for easy and fast gesture-to-sound scenarios prototyping. It includes receivers from motion capture systems, signal processing modules, machine learning modules, and sound synthesis modules. Overview patch on the right.

Github repo:
See also the academic paper