info:eu-repo/semantics/article
A Bayesian approach to convolutive nonnegative matrix factorization for blind speech dereverberation
Fecha
2018-10Registro en:
Ibarrola, Francisco Javier; Di Persia, Leandro Ezequiel; Spies, Ruben Daniel; A Bayesian approach to convolutive nonnegative matrix factorization for blind speech dereverberation; Elsevier Science; Signal Processing; 151; 10-2018; 89-98
0165-1684
CONICET Digital
CONICET
Autor
Ibarrola, Francisco Javier
Di Persia, Leandro Ezequiel
Spies, Ruben Daniel
Resumen
When a signal is recorded in an enclosed room, it typically gets affected by reverberation. This degradation represents a problem when dealing with audio signals, particularly in the field of speech signal processing, such as automatic speech recognition. Although there are some approaches to deal with this issue that are quite satisfactory under certain conditions, constructing a method that works well in a general context still poses a significant challenge. In this article, we propose a Bayesian approach based on convolutive nonnegative matrix factorization that uses prior distributions in order to impose certain characteristics over the time-frequency components of the restored signal and the reverberant components. An algorithm for implementing the method is described and tested.Comparisons of the results against those obtained with state-of-the-art methods are presented, showing significant improvement.