Austria

Emmanouil Karystinaios

Advanced
@manoskary

Postdoctoral Research in AI and Music at the Computational Perception Institute of Johannes Kepler University.

emmanouil-karystinaios.github.io/

graphmuse. A Graph Deep Learning Library for Music.

113

weavemuse. An open agentic system built on smolagents, integrating multimodal state-of-the-art music AI models for understanding, generation, and interaction.

32

MusGConv. Implementation and experiment of the MusGConv paper.

15

ChordGNN. This is the repository of the paper: Roman Numeral Analysis with Graph Neural Networks

13

Steer-SAO. Adaptors for Stable-Audio-3 controlled generation.

12

SMUG-Explain. A Framework for Symbolic MUsic Graph Explanations

11

analysisgnn. A Unified Music Analysis Model with Graph Neural Networks

10

vocsep_ijcai2023. Code for the paper Musical Voice Separation as Link Prediction: Modeling a Musical Perception Task as a Multi-Trajectory Tracking Problem

9

cadet. Cadence Detection in Symbolic Classical Music using Graph Neural Networks

7

musym-GDL. Geometric Deep Learning Applied on Symbolic Music Scores

6

audio-enhancement. Pipeline for Realtime Music Enhancement

5

tonnetzcad. A Data-Set for GDL containing heterogeneous graphs produced by Tonnetz trajectories with cadence labels.

2

Topological-Descriptors-for-Symbolic-Music-Genre-Classification. During the period of November 2019 and August 2020 I carried out my internship atT ́el ́ecom Paris in Paris. I worked under the Supervision of Isabelle Block as part of theImages team. We overtook the endeavour of automated musical analysis for musical genreclassification. The goal was to refine the notion of harmonic trajectory descriptor andinvestigate how to improve classification on symbolic music data. The hypothesis is thatthe harmonic trajectory of musical piece is an imprint that defines the genre of the pieceat a certain level.In this work, we focus on different methods to utilize the harmonic trajectory descriptoreither for classification purposes or transcription and key estimation purposes. We workedwith large databases of symbolic music scores, such as the Lakh Data-set. We presented amodular, in regards to descriptors, system for supervised learning, mainly using supportvector machines.

2
13
Apply