This paper addresses the problem of discriminative training of language models that does not require any transcribed acoustic data. We propose to minimize the conditional entropy ...
Abstract. In this paper, we propose a method for blind source separation (BSS) of convolutive audio recordings with short blocks of stationary sources, i.e. dynamically changing so...
A vertex-cut X is said to be a restricted cut of a graph G if it is a vertex-cut such that no vertex u in G has all its neighbors in X. Clearly, each connected component of G - X ...
We discuss adaptive strategies for choosing regularization parameters in TikhonovPhillips regularization of discretized linear operator equations. Two rules turn out to be entirel...
Abstract. We consider the problem of learning a user's ordinal preferences on a multiattribute domain, assuming that her preferences are lexicographic. We introduce a general ...