Inferring Score Level Musical Information From Low-Level Musical Data

This thesis is focussed on inferring score level information from low-level symbolic musical data. It gives an introduction into approaches for musical quantisation, tempo detection, voice separation, as well as key and meter detection or the inference of secondary score level information. The thesis introduces also new approaches based on pattern detection and statistical analysis for the core issues tempo detection and quantisation. For the voice separation problem the thesis introduces a now approach based on local search.