About Me

I am a researcher in computer science, data science and machine learning, with a focus on intelligent music processing. I have received a Bachelor’s degree from the University of Vienna, a Master’s degree from the Technical University of Vienna, and a PhD from Johannes Kepler University in Linz, Austria.

Currently, I am a machine learning engineer at Apple. Before that, I was an assistant professor (“Universitätsassistent”) at the Institute for Computational Perception at the Johannes Kepler University in Linz, Austria. I enjoy research in a variety of topics (see below), as well as teaching and supervising students, and supporting them in their endeavours. In addition, it is my utmost concern to positively influence the working environment, and organise activities and formal and informal meetings at our institute.

Research Interests

Real-time Music Tracking

Algorithms that ‘listen’ to a live performance of a musician and ‘read’ along in the score, i.e. at any time compute the current position; feature representations, tempo models, multi-agent tracking, robustness and flexibility considerations; results have been shown live at various occasions (including at a concert at the venerable Concertgebouw in Amsterdam).

Music Synchronisation

Alignment of multiple representations of a piece of music (i.e. symbolic representations, audio recordings of performances, the sheet music, etc.); data collection, storage and annotation; (semi-)automatic methods (from classic alignment algorithms to deep learning) that try to solve multi-modal synchronisation tasks with minimal human intervention.

Machine Learning and Music.

Multi-modal machine learning in the context of music (audio, video, motion, sheet music images, symbolic scores); Learning musically-motivated mid-level features, for example transposition-invariant audio features via gated autoencoders; Reinforcement learning in the context of intelligent music processing.

Music Identification

Identification of a piece of music, given a short audio query; tempo- and transposition-invariant fingerprinting methods; real-time identification scenarios.

Live Demonstrations and Interaction

Development of applications and demonstration prototypes; flexible piano music companion; live tracking and music visualisation in concert halls.

Miscellaneous Research Topics

Pattern recognition in classical music, onset detection, beat and rhythm tracking, music transcription, music representations, automatic accompaniment and music visualisations.