Projects

Selected tools and research implementations in computational musicology, symbolic score analysis, and audio-based machine learning.

Research Software

• CAMAT: Cloud-oriented toolbox for symbolic score analysis with MEI and MusicXML workflows. Explore CAMAT
• AudioSpylt: Python toolkit for audio feature extraction and interactive visualisation. View on GitHub
• FFTimbre: Notebooks and utilities for FM/additive timbre matching and evaluation. View on GitHub
• beat_it: Research notebooks for beat and tempo fluctuation analysis. View on GitHub

Methods and Stack

• Scientific Python: NumPy, SciPy, pandas, scikit-learn, Jupyter, music21, librosa.
• Machine learning: PyTorch-based prototyping for timbre, rhythm, and structure analysis.
• Research engineering: reproducible notebooks, Git/GitHub workflows, and cloud-oriented deployment with Jupyter infrastructures.

For full project context, publications, and talks, see the Curriculum Vitae.