MULTI-AGENT MULTIMODAL PERFORMANCE ANALYSIS
Arne Eigenfeldt, Ajay Kapur · 2008
This paper describes a custom built system which extracts high-level musical information from real-time multimodal gesture data. Data is collected from sensors during rehearsal using one program, GATHER, and, combined with audio analysis, is used to derive statistical coefficients which are used to identify three different playing styles of North Indian music: Alap, Ghat, and Jhala. A real-time program, LISTEN, uses these coefficients in a multi-agent analysis to determine the current playing style. 1.