Master 2017 2018
Stages de la spécialité SAR
Deep Learning for Polyphonic Music Generation

The student will be involved in developing novel generative music architecture and systems based on neural network architecture developed at the Metacreation Lab. While style imitation automate the composition / style imitation task at human competitive level, a lot of work remains for these systems to become useful computer-assisted composition aides. This internship will leverage the applicants skills, to contribute to the overall project or any of its parts (data augmentation, corpus management, multi-corpus integration, output evaluation, ...). There is also opportunities in implementing non Neural-network based approached for generation for the sake of comparison.

Bibliographie

An Introduction to Musical Metacreation, Philippe Pasquier, Oliver Bown, Arne Eigenfeldt, Schlomo Dubnov ACM Computers in Entertainment, 2016.

Deep Learning Techniques for Music Generation - A Survey, Jean-Pierre Briot, Gaëtan Hadjeres, François Pachet, https://arxiv.org/abs/1709.01620