Real-Time Generative Music: A Study of Deep-Learning–Ready Techniques

This internship focuses on the study of deep learning approaches applied to real-time generative music. It aims to analyze existing methods that enable the instantaneous production of coherent and expressive musical content responsive to interactions or temporal constraints. The work involves identifying neural architectures and learning strategies best suited for the continuous generation of audio or symbolic streams, while assessing their performance, expressivity, and creative potential. The project lies at the intersection of artificial intelligence, audio signal processing, and computer-assisted composition.