Fast Estimation of Optimal Sparseness of Music Signals

A. la Cour-Harbo (Denmark)


sparse representations, matching pursuit, alternating projection, music.


We want to use a variety of sparseness measured applied to ‘the minimal 1 norm representation’ of a music signal in an over-complete dictionary as features for automatic clas sification of music. Unfortunately, the process of comput ing the optimal 1 norm representation is rather slow, and we therefore investigate the use of matching pursuit, alter nating projection, and Moore-Penrose inverse for estimat ing the result of applying two different sparseness measures to ‘the minimal 1 norm representation’ without actually computing this representation.

Important Links:

Go Back