We propose a novel algorithm for the separation of convolutive speech mixtures using two-microphone recordings, based on the combination of independent component analysis (ICA) an...
Most existing binaural approaches to speech segregation rely on spatial filtering. In environments with minimal reverberation and when sources are well separated in space, spatial...
John Woodruff, Rohit Prabhavalkar, Eric Fosler-Lus...
We show how a preselection of hidden variables can be used to efficiently train generative models with binary hidden variables. The approach is based on Expectation Maximization (...
This paper suggests a framework for mining subjectively interesting pattern sets that is based on two components: (1) the encoding of prior information in a model for the data min...
Tijl De Bie, Kleanthis-Nikolaos Kontonasios, Eirin...
This work focuses on developing systems of blocks in SIMULINK and VHDL to reuse on design of applications involving the recognition of polygonal objects. Usage of this work reduce...