Improvasher: A Real-Time Mashup System For Live Musical Input
Matthew E. P. Davies, Adam M. Stark, Fabien Gouyon, Masataka Goto · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · 2014
In this paper we present Improvasher a real-time musical accompaniment system which creates an automatic mashup to accompany live musical input. Improvasher is built around two music processing modules, the first, a performance following technique, makes beat-synchronous predictions of chroma features from a live musical input. The second, a music mashup system, determines the compatibility between beat-synchronous chromagrams from different pieces of music. Through the combination of these two techniques, a real-time time predict mashup can be generated towards a new form of automatic accompaniment for interactive musical performance.