Ha-Hi-Hun: Performance Rendering System of High Controllability
Keiji Hirata and Rumi Hiraga
This paper presents the design principle and algorithm of a new performance rendering framework, and a prototype system based on the framework. We believe that a practical performance rendering system should be able to refine and improve a generated performance interactively, incrementally, and locally through direct instructions issued by a user. In addition, the generated performance must reflect the user's intention properly. For these purposes, we propose a new framework called two-stage performance rendering. The first stage translates a user's instruction into the deviations of the onset time, duration, and amplitude of structurally important notes. The second stage spreads the deviations over surrounding notes. Lastly, we introduce a performance rendering system called ``Ha-Hi-Hun'' that we are developing.