Stefan Balke

Hello, I’m Stefan. I work at the intersection of music, technology, and community: researching Music Information Retrieval (MIR), building datasets and demonstrators for musical audio, and creating cultural projects with local impact.

I completed my PhD in MIR in 2017 in Meinard Müller’s group at AudioLabs Erlangen. After a PostDoc with Gerhard Widmer’s group at the Institute for Computational Perception at JKU Linz, I worked for three years as a Data Scientist and Team Lead in consulting. In the winter term of 2023/24, I taught as a professor at Hochschule Weserbergland in Hameln, Germany.

Currently, I am back at AudioLabs Erlangen as a PostDoc in Meinard Müller’s group, where my research focuses on wind music, concert band repertoire, and computational tools for understanding and working with musical audio. Beyond research, I am active as a trumpet player, conductor, board member, and project initiator in the amateur music scene in and around Höxter.

A detailed version of my curriculum vitae is available on request.

news

Aug 31, 2026	Teaching Selected Topics in Deep Learning for Audio, Speech, and Music Processing in the winter term 2026/2027.
Jun 30, 2026	The WindMusicLab concert brought our research on wind music to the stage in Höxter. 🎺
Jun 30, 2026	Call for papers for the TISMIR Special Collection on Open Music Data for Music Processing Research.
Feb 28, 2026	A donation from Round Table 150 Höxter-Holzminden helped us purchase new instruments for the primary-school wind class. 🎺
Jan 31, 2026	Our TISMIR paper on the re-release of the RWC Music Database was published.

selected publications

ChoraleBricks: A Modular Multitrack Dataset for Wind Music Research

Stefan Balke, Axel Berndt, and Meinard Müller

Transaction of the International Society for Music Information Retrieval (TISMIR), 2025

DOI Bib Code

@article{BalkeBM25_ChoraleBricks_TISMIR,
  author = {Balke, Stefan and Berndt, Axel and M{\"u}ller, Meinard},
  title = {{ChoraleBricks}: {A} Modular Multitrack Dataset for Wind Music Research},
  journal = {Transaction of the International Society for Music Information Retrieval ({TISMIR})},
  volume = {8},
  number = {1},
  pages = {39--54},
  year = {2025},
  doi = {10.5334/tismir.252},
  url-demo = {https://www.audiolabs-erlangen.de/resources/MIR/2025-ChoraleBricks},
  url-details = {https://transactions.ismir.net/articles/10.5334/tismir.252},
}

Score Following as a Multi-Modal Reinforcement Learning Problem

Florian Henkel, Stefan Balke, Matthias Dorfer, and Gerhard Widmer

Transactions of the International Society for Music Information Retrieval (TISMIR), 2019

DOI Bib PDF Code

@article{HenkelBDW19_ScoFoRL_TISMIR,
  author = {Henkel, Florian and Balke, Stefan and Dorfer, Matthias and Widmer, Gerhard},
  title = {Score Following as a Multi-Modal Reinforcement Learning Problem},
  journal = {Transactions of the International Society for Music Information Retrieval ({TISMIR})},
  volume = {2},
  number = {1},
  pages = {67--81},
  year = {2019},
  doi = {http://doi.org/10.5334/tismir.31},
}

Cross-Modal Music Retrieval and Applications: An Overview of Key Methodologies

Meinard Müller, Andreas Arzt, Stefan Balke, Matthias Dorfer, and Gerhard Widmer

IEEE Signal Processing Magazine, 2019

DOI Bib PDF

@article{MuellerABDW19_MusicRetrieval_IEEE-SPM,
  author = {M{\"u}ller, Meinard and Arzt, Andreas and Balke, Stefan and Dorfer, Matthias and Widmer, Gerhard},
  title = {Cross-Modal Music Retrieval and Applications: An Overview of Key Methodologies},
  journal = {{IEEE} Signal Processing Magazine},
  volume = {36},
  number = {1},
  pages = {52--62},
  year = {2019},
  url = {https://doi.org/10.1109/MSP.2018.2868887},
  doi = {10.1109/MSP.2018.2868887},
}