Precise Pitch Control in Real Time Corpus-Based Concatenative Synthesis

The need for fine-tuned microtonal pitch combined with the timbral richness of corpus-based concatenative synthesis has led to the development of a new tool for corpus-based pitch and loudness control in real time with CATART. Drawing on recent research in feature modulation synthesis (FMS) as well as the bach library for MAX/MSP, we have implemented a set of new modules for CATART that permit the user to define microtonal har- monies graphically and combine them with other audio descriptors to trigger concatenative synthesis in real or deferred time. Pitch information is generated from a pitch analysis or extracted from soundfile meta-data, and loudness may be controlled independently for different sound sets. Musical implementations already suggest promising results as well as future goals to generalize this approach to further timbral features for corpus-based FMS.