Exploring Musical Possibilities via Machine Simulation

Informations projet

Whither Music?

N° de convention de subvention: 101019375

DOI

10.3030/101019375

Date de signature de la CE 28 Avril 2021

Date de début 1 Janvier 2022

Date de fin 31 Decembre 2027

Financé au titre de

• EXCELLENT SCIENCE - European Research Council (ERC)

Coût total

€ 2 500 000,00

Contribution de l’UE

€ 2 500 000,00

Coordonné par

UNIVERSITAT LINZ
Austria

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Publications

Conference proceedings (49)

Efficient Large-scale Audio Tagging via Transformer-to-CNN Knowledge Distillation

Auteurs: Florian Schmid, Khaled Koutini, Gerhard Widmer
Publié dans: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2023, 2023
Éditeur: IEEE
DOI: 10.48550/arxiv.2211.04772

The match File Format: Encoding Alignments between Scores and Performances

Auteurs: Francesco Foscarin, Emmanouil Karystinaios, Silvan David Peter, Carlos Cancino-Chacón, Maarten Grachten, Gerhard Widmer
Publié dans: Proceedings of the Music Encoding Conference (MEC 2022), 2022
Éditeur: Tne Music Encoding Initiative (MEI)
DOI: 10.48550/arxiv.2206.01104

8+8=4: Formalizing Time Units to Handle Symbolic Music Durations

Auteurs: Emmanouil Karystinaios, Francesco Foscarin, Florent Jacquemard, Masahiko Sakai, Satoshi Tojo, Gerhard Widmer
Publié dans: Proceedings of the 16th International Symposium on Computer Music Multidisciplinary Research (CMMR 2023), 2023
Éditeur: Springer
DOI: 10.48550/arxiv.2310.14952

Cluster and Separate: A GNN Approach to Voice and Staff Prediction for Score Engraving

Auteurs: Francesco Foscarin, Emmanouil Karystinaios, Eita Nakamura, Gerhard Widmer
Publié dans: Proceedings of the 25th International Society for Music Information Retrieval Conference (ISMIR 2024), 2024
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2407.21030

The ACCompanion: Combining Reactivity, Robustness, and Musical Expressivity in an Automatic Piano Accompanist

Auteurs: Carlos Cancino-Chacón, Silvan Peter, Patricia Hu, Emmanouil Karystinaios, Florian Henkel, Francesco Foscarin, Nimrod Varga, Gerhard Widmer
Publié dans: Proceedings of the 32nd International Joint Conference on Artificial Intelligence (IJCAI-23), 2023
Éditeur: IJCAI
DOI: 10.48550/arxiv.2304.12939

EngravingGNN: A Hybrid Graph Neural Network for End-to-End Piano Score Engraving

Auteurs: Emmanouil Karystinaios, Francesco Foscarin, Gerhard Widmer
Publié dans: Proceedings of the International Conference on Technologies for Music Notation and Representation (TENOR), 2025
Éditeur: Central Conservatory of Music Beijing
DOI: 10.48550/arxiv.2509.19412

Are we describing the same sound? An Analysis of Word Embedding Spaces of Expressive Piano Performance

Auteurs: Silvan Peter, Shreyan Chowdhury, Carlos Cancino-Chacón, Gerhard Widmer
Publié dans: Proceedings of the Forum for Information Retrieval Evaluation (FIRE 2023), 2023
Éditeur: Association for Computing Machinery (ACM)
DOI: 10.1145/3632754.3632759

Passage Summarization with Recurrent Models for Audio-Sheet Music Retrieval

Auteurs: Luis Carvalho, Gerhard Widmer
Publié dans: Proceedings of the 24th Conference of the International Society for Music Information Retrieval (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2309.12111

Music Boomerang: Reusing Diffusion Models for Data Augmentation and Audio Manipulation

Auteurs: Alexander Fichtinger, Jan Schlüter, Gerhard Widmer
Publié dans: Proceedings of the 22nd Sound and Music Computing Conference (SMC 2025), 2025
Éditeur: Sound and Music Computing Initiative
DOI: 10.48550/arxiv.2507.04864

The GlueNote: Learned Representations for Robust and Flexible Note Alignment

Auteurs: Silvan Peter, Gerhard Widmer
Publié dans: Proceedings of the 25th International Society for Music Information Retrieval Conference (ISMIR 2024), 2024
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2408.04309

Are Inherently Interpretable Models More Robust? A Study in Music Emotion Recognition

Auteurs: Katharina Hoedt, Arthur Flexer, Gerhard Widmer
Publié dans: Proceedings of the 22nd Sound and Music Computing Conference (SMC 2025), 2025
Éditeur: Sound and Music Computing Initiative
DOI: 10.5281/zenodo.15837274

Partitura: A Python Package for Symbolic Music Processing

Auteurs: Carlos Cancino-Chacón, Silvan David Peter, Emmanouil Karystinaios, Francesco Foscarin, Maarten Grachten, Gerhard Widmer
Publié dans: Proceedings of the Music Encoding Conference (MEC 2022), 2022
Éditeur: The Music Encoding Initiative (MEI)
DOI: 10.48550/arxiv.2206.01071

Estimating Musical Surprisal from Audio in Autoregressive Diffusion Model Noise Spaces

Auteurs: Mathias Rose Bjare, Stefan Lattner, Gerhard Widmer
Publié dans: Proceedings of the 26th International Society for Music Information Retrieval Conference (ISMIR), 2025
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2508.05306

The Rach3 Dataset: Towards Data-Driven Analysis of Piano Performance Rehearsal

Auteurs: Carlos Cancino-Chacón, Ivan Pilkov
Publié dans: Proceedings of the 30th International Conference on Multimedia Modeling (MMM 2024), 2024
Éditeur: Springer Verlag
DOI: 10.1007/978-3-031-56435-2_3

Decoding and Visualising Intended Emotion in an Expressive Piano Performance

Auteurs: Shreyan Chowdhury, Gerhard Widmer
Publié dans: 23rd International Society for Music Information Retrieval Conference (ISMIR 2022), Late-Breaking/Demo Papers, 2022
Éditeur: International Society for Music Information Retrieval (ISMIR)
DOI: 10.48550/arxiv.2303.01875

Beat this: Accurate Beat Tracking Without DBN Postprocessing

Auteurs: Francesco Foscarin, Jan Schlüter, Gerhard Widmer
Publié dans: Proceedings of the 25th International Society for Music Information Retrieval Conference (ISMIR 2024), 2024
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2407.21658

AnalysisGNN: Unified Music Analysis with Graph Neural Networks

Auteurs: Emmanouil Karystinaios, Johannes Hentschel, Markus Neuwirth, Gerhard Widmer
Publié dans: 17th International Symposium on Computer Music Multidisciplinary Research (CMMR), 2025
Éditeur: University College London
DOI: 10.48550/arxiv.2509.06654

Sounding Out Reconstruction Error-Based Evaluation of Generative Models of Expressive Performance

Auteurs: Silvan Peter, Carlos Cancino-Chacón, Emmanouil Karystinaios, Gerhard Widmer
Publié dans: Proceedings of the 10th International Conference on Digital Libraries for Musicology (DLfM '23), 2023, Page(s) 58-66
Éditeur: ACM
DOI: 10.1145/3625135.3625141

Discrete Diffusion Probabilistic Models for Symbolic Music Generation

Auteurs: Matthias Plasser, Silvan Peter, Gerhard Widmer
Publié dans: Proceedings of the 32nd International Joint Conference on Artificial Intelligence (IJCAI-23), 2023
Éditeur: IJCAI

Exploring System Adaptations For Minimum Latency Real-Time Piano Transcription

Auteurs: Patricia Hu, Silvan David Peter, Jan Schlüter, Gerhard Widmer
Publié dans: Proceedings of the 26th International Society for Music Information Retrieval Conference (ISMIR 2025), 2025
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2509.07586

The Batik-plays-Mozart Corpus: Linking Performance to Score to Musicological Annotations

Auteurs: Patricia Hu, Gerhard Widmer
Publié dans: Proceedings of the 24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2309.02399

"Concept-Based Techniques for ""Musicologist-friendly"" Explanations in a Deep Music Classifier"

Auteurs: Francesco Foscarin, Katharina Hoedt, Verena Praher, Arthur Flexer, Gerhard Widmer
Publié dans: Proceedings of the 23rd International Society for Music Information Retrieval Conference (ISMIR 2022), 2022
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2208.12485

Estimating Musical Surprisal in Audio

Auteurs: Mathias Rose Bjare, Giorgia Cantisani, Stefan Lattner, Gerhard Widmer
Publié dans: Proceedings of the 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2025), 2025
Éditeur: IEEE
DOI: 10.48550/arxiv.2501.07474

Exploring Sampling Techniques for Generating Melodies with a Transformer Language Model

Auteurs: Mathias Rose Bjare, Stefan Lattner, Gerhard Widmer
Publié dans: Proceedings of the 24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2308.09454

Perception-Inspired Graph Convolution for Music Understanding Tasks

Auteurs: Emmanouil Karystinaios, Francesco Foscarin, Gerhard Widmer
Publié dans: Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI-24), 2024
Éditeur: International Joint Conferences on Artificial Intelligence, Inc. (IJCAI)
DOI: 10.48550/arxiv.2405.09224

Cadence Detection in Symbolic Classical Music using Graph Neural Networks

Auteurs: Emmanouil Karystinaios and Gerhard Widmer
Publié dans: Proceedings of the 23rd International Society for Music Information Retrieval Conference (ISMIR 2022), 2022
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2208.14819

Towards Robust and Truly Large-Scale Audio-Sheet Music Retrieval

Auteurs: Luis Carvalho, Gerhard Widmer
Publié dans: Proceedings of the IEEE 6th International Conference on Multimedia Information Processing and Retrieval (MIPR), 2023
Éditeur: IEEE
DOI: 10.48550/arxiv.2309.12158

SMUG-Explain: A Framework for Symbolic Music Graph Explanations

Auteurs: Emmanouil Karystinaios, Francesco Foscarin, Gerhard Widmer
Publié dans: Proceedings of the Sound and Music Computing Conference 2024 (SMC2024), 2024
Éditeur: Sound and Music Computing Network
DOI: 10.48550/arxiv.2405.09241

Predicting Music Hierarchies with a Graph-Based Neural Decoder

Auteurs: Francesco Foscarin, Daniel Harasim, Gerhard Widmer
Publié dans: Proceedings of the 24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2306.16955

Self-Supervised Contrastive Learning for Robust Audio-Sheet Music Retrieval Systems

Auteurs: Luis Carvalho, Tobias Washüttl, Gerhard Widmer
Publié dans: Proceedings of the 14th ACM Conference on Multimedia Systems (MMSys 2023), 2023
Éditeur: Association for Computing Machinery (ACM)
DOI: 10.1145/3587819.3590968

How to Infer Repeat Structures in MIDI Performances

Auteurs: Silvan Peter, Patricia Hu, Gerhard Widmer
Publié dans: Proceedings of the Music Encoding Conference (MEC 2025), 2025
Éditeur: Music Encoding Initiative
DOI: 10.48550/arxiv.2505.05055

Differentiable Dictionary Search: Integrating Linear Mixing with Deep Non-Linear Modelling for Audio Source Separation

Auteurs: Lukas Samuel Martak, Rainer Kelz, Gerhard Widmer
Publié dans: Proceedings of the 24th International Congress on Acoustics (ICA 2022), 2022
Éditeur: International Commission for Acoustics (ICA)

Defending a Music Recommender Against Hubness-Based Adversarial Attacks

Auteurs: Katharina Hoedt, Arthur Flexer, Gerhard Widmer
Publié dans: Proceedings of the 19th Sound and Music Computing Conference (SMC 2022), 2022
Éditeur: Telecom Saint-Etienne
DOI: 10.5281/zenodo.6573391

Symbolic Music Representations for Classification Tasks: A Systematic Evaluation

Auteurs: Huan Zhang, Emmanouil Karystinaios, Simon Dixon, Gerhard Widmer, Carlos Eduardo Cancino-Chacón
Publié dans: Proceedings of the 24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2309.02567

GraphMuse: A Library for Symbolic Music Graph Processing

Auteurs: Emmanouil Karystinaios, Gerhard Widmer
Publié dans: Proceedings of the 25th International Society for Music Information Retrieval Conference (ISMIR 2024), 2024
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2407.12671

Pairing Real-Time Piano Transcription with Symbol-level Tracking for Precise and Robust Score Following

Auteurs: Silvan Peter, Patricia Hu, Gerhard Widmer
Publié dans: Proceedings of the 22nd Sound and Music Computing Conference (SMC 2025), 2025
Éditeur: Sound and Music Computing Initiative
DOI: 10.48550/arxiv.2505.05078

Optical Music Recognition of Jazz Lead Sheets

Auteurs: Juan Carlos Martinez-Sevilla, Francesco Foscarin, Patricia Garcia-Iasci, David Rizo, Jorge Calvo-Zaragoza, Gerhard Widmer
Publié dans: Proceedings of the 26th International Society for Music Information Retrieval Conference (ISMIR 2025), 2025
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2509.05329

Towards Musically Informed Evaluation of Piano Transcription Models

Auteurs: Patricia Hu, Lukas Martak, Carlos Cancino-Chacon, Gerhard Widmer
Publié dans: Proceedings of the 25th International Society for Music Information Retrieval Conference (ISMIR 2024), 2024
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2406.08454

Exploring Performance-Complexity Trade-Offs in Sound Event Detection Models

Auteurs: Tobias Morocutti, Florian Schmid, Jonathan Greif, Francesco Foscarin, Gerhard Widmer
Publié dans: Proceedings of the 33rd European Signal Processing Conference (EUSIPCO 2025), 2025
Éditeur: IEEE
DOI: 10.48550/arxiv.2503.11373

Online Symbolic Music Alignment with Offline Reinforcement Learning

Auteurs: Silvan Peter
Publié dans: Proceedings of the 24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2401.00466

Controlling Surprisal in Music Generation via Information Content Curve Matching

Auteurs: Mathias Bjare, Stefan Lattner, Gerhard Widmer
Publié dans: Proceedings of the 25th International Society for Music Information Retrieval Conference (ISMIR 2024), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2408.06022

Learning General Audio Representations with Large-Scale Training of Patchout Audio Transformers

Auteurs: Khaled Koutini, Shahed Masoudian, Florian Schmid, Hamid Eghbal-zadeh, Jan Schlüter, Gerhard Widmer
Publié dans: Proceedings of HEAR 2021: Holistic Evaluation of Audio Representations, 2022, Page(s) PMLR 166, 65-89, ISSN 2640-3498
Éditeur: JMLR, Inc.
DOI: 10.48550/arxiv.2211.13956

Music Visualisation and Its Short-Term Effect on Appraisal Skills

Auteurs: Ali Nikrang, Maarten Grachten, Martin Gasser, Harald Frostel, Gerhard Widmer, Tom Collins
Publié dans: HCI International 2023 – Late Breaking Papers. HCII 2023, 2023
Éditeur: Springer
DOI: 10.1007/978-3-031-48044-7_9

Musical Voice Separation as Link Prediction: Modeling a Musical Perception Task as a Multi-Trajectory Tracking Problem

Auteurs: Emmanouil Karystinaios, Francesco Foscarin, Gerhard Widmer
Publié dans: Proceedings of the 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023), 2023
Éditeur: IJCAI

Roman Numeral Analysis with Graph Neural Networks: Onset-wise Predictions from Note-wise Features

Auteurs: Emmanouil Karystinaios, Gerhard Widmer
Publié dans: Proceedings of the 24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2307.03544

Language Models for Music Medicine Generation

Auteurs: Emmanouil Nikolakakis, Joann Ching, Emmanouil Karystinaios, Gabrielle Sipin, Gerhard Widmer, Razvan Marinescu
Publié dans: Extended Abstracts, Late-Breaking Demo Session, 25th International Society for Music Information Retrieval Conference (ISMIR 2024),, 2024
Éditeur: International Society for Music Information Retrieval
DOI: 10.48550/arxiv.2411.09080

Quantifying the Corpus Bias Problem in Automatic Music Transcription Systems

Auteurs: Lukas Samuel Martak, Patricia Hu, Gerhard Widmer
Publié dans: 1st International Workshop on Sound Signal Processing Applications (IWSSPA), 2024
Éditeur: CMMSE
DOI: 10.48550/arxiv.2408.04737

Concept-based Explanations for Music Emotion Recognition

Auteurs: Verena Praher, Verena Szojak, Gerhard Widmer
Publié dans: Proceedings of the 22nd Sound and Music Computing Conference (SMC 2025), 2025
Éditeur: Sound and Music Computing Initiative
DOI: 10.5281/zenodo.15837275

Expressivity-aware Music Performance Retrieval using Mid-level Perceptual Features and Emotion Word Embeddings

Auteurs: Shreyan Chowdhury, Gerhard Widmer
Publié dans: Proceedings of the Forum for Information Retrieval Evaluation Conference /(FIRE 2023), 2023
Éditeur: ACM

Peer reviewed articles (7)

Commentary on “A Computational Approach to the Detection and Prediction of (Ir)Regularity in Children’s Folk Songs”

Auteurs: Carlos Cancino-Chacón
Publié dans: Empirical Musicology Review, Numéro 16(2), 2023, Page(s) 328-335, ISSN 1559-5749
Éditeur: The Ohio State University Libraries
DOI: 10.18061/emr.v16i2.9159

Differentiable Short-Term Models for Efficient Online Learning and Prediction in Monophonic Music

Auteurs: Mathias Rose Bjare, Stefan Lattner, Gerhard Widmer
Publié dans: Transactions of the International Society for Music Information Retrieval, Numéro 5(1), 2022, Page(s) 190-207, ISSN 2514-3298
Éditeur: Ubiquity Press
DOI: 10.5334/tismir.123

Automatic Note-Level Score-to- Performance Alignments in the ASAP Dataset

Auteurs: Silvan Peter, Carlos Eduardo Cancino-Chacón, Francesco Foscarin, Andrew P. McLeod, Florian Henkel, Emmanouil Karystinaios, Gerhard Widmer
Publié dans: Transactions of the International Society for Music Information Retrieval, Numéro 6(1), 2023, Page(s) 27–42, ISSN 2514-3298
Éditeur: Ubiquity Press
DOI: 10.5334/tismir.149

Balancing Bias and Performance in Polyphonic Piano Transcription Systems

Auteurs: Lukas Samuel Martak, Rainer Kelz, Gerhard Widmer
Publié dans: Frontiers in Signal Processing, Numéro 2:975932, 2022, ISSN 2673-8198
Éditeur: Frontiers Media SA
DOI: 10.3389/frsip.2022.975932

DExter: Learning and Controlling Performance Expression with Diffusion Models

Auteurs: Huan Zhang, Shreyan Chowdhury, Carlos Eduardo Cancino-Chacón, Jinhua Liang, Simon Dixon, Gerhard Widmer
Publié dans: Applied Sciences, Numéro 14(15), 2024, Page(s) 1-17, ISSN 2076-3417
Éditeur: MDPI
DOI: 10.3390/app14156543

Dynamic Convolutional Neural Networks as Efficient Pre-Trained Audio Models

Auteurs: Florian Schmid, Khaled Koutini, Gerhard Widmer
Publié dans: IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2024, ISSN 2329-9290
Éditeur: IEEE Advancing Technology for Humanity
DOI: 10.1109/taslp.2024.3376984

Constructing Adversarial Examples to Investigate the Plausibility of Explanations in Deep Audio and Image Classifiers

Auteurs: Katharina Hoedt, Verena Praher, Arthur Flexer, Gerhard Widmer
Publié dans: Neural Computing and Applications, Numéro 2022, 2022, ISSN 0941-0643
Éditeur: Springer Verlag
DOI: 10.1007/s00521-022-07918-7

Thesis and dissertations (7)

The Human Performance: Computational Methods for the Analysis of Played Classical Piano Music

Auteurs: Silvan Peter
Publié dans: 2025
Éditeur: Johannes Kepler University Linz

Symbolic Music Generation using Discrete Diffusion Probabilistic Models

Auteurs: Matthias Plasser
Publié dans: Master Thesis, 2023
Éditeur: Johannes Kepler University Linz (JKU)

(de)Fusion of Score and Performance Representations for Symbolic Music Generation

Auteurs: Tara Jadidi
Publié dans: Master Thesis, 2024
Éditeur: Johannes Kepler University (JKU)

Multi-modal Deep Learning for On-line Music Following in Score Sheet Images

Auteurs: Florian Henkel
Publié dans: PhD Thesis, 2022
Éditeur: Johannes Kepler University Linz

Multi-modal Music Search and Retrieval Without Symbolic Representations

Auteurs: Luis Carvalho
Publié dans: PhD Thesis, 2025
Éditeur: Johannes Kepler University

Modelling Emotional Expression in Music Using Interpretable and Transferable Perceptual Features

Auteurs: Shreyan Chowdhury
Publié dans: PhD Thesis, 2022
Éditeur: Johannes Kepler University Linz

Symbolic Music Analysis with Graph Neural Networks

Auteurs: Emmanouil Karystinaios
Publié dans: Ph.D. Thesis, 2024
Éditeur: Johannes Kepler University (JKU)