Emmanouil Karystinaios

Emmanouil Karystinaios is a Post-Doctoral Researcher at the Institute of Computational Perception, Johannes Kepler University Linz, Austria, and holds concurrent Lecturer roles at Kunstuniversität Linz and Universität für Musik und darstellende Kunst Wien. His research focuses on AI applications in music, particularly leveraging graph neural networks (GNNs) for symbolic music analysis, optical music recognition, and generative music medicine. He leads projects such as GraphMuse, a Python library for symbolic music graph processing, and SMUG-Explain, an explainability framework for music data.

Education includes a PhD in Computational Perception (JKU Linz, ongoing since 2020), MSc in Mathematical Logic (Université Paris Diderot), and MA in Musical Programming/Composition (Université Paris Vincennes). His work bridges computational methods with music theory, emphasizing explainable AI and music perception tasks.

Recent publications highlight contributions to voice separation via GNNs, score engraving optimization, and music medicine using language models. His research also addresses expressive performance evaluation and symbolic music representation challenges, with applications in automated accompaniment systems like the ACCompanion.