Linear program differentiation for single-channel speech separation
B.A. Pearlmutter and R. K. Olsson
In: MLSP, Maynooth, Ireland(2006).

Abstract

Many apparently difficult problems can be solved by reduction to linear programming. Such problems are often subproblems within larger systems. When gradient optimisation of the entire larger system is desired, it is necessary to propagate gradients through the internally-invoked LP solver. Here we apply it to gradient optimisation of overcomplete dictionaries for maximally sparse representations of a speech corpus. The dictionaries are employed in a single-channel speech separation task, leading to \unit[5]{dB} and \unit[8]{dB} target-to-interference ratio improvements for same-gender and opposite-gender mixtures, respectively. Furthermore, the dictionaries are successfully applied to a speaker identification task.

[Edit]