13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

PLDA using Gaussian Restricted Boltzmann Machines with application to Speaker Verification

Themos Stafylakis (1,2), Patrick Kenny (1), Mohammed Senoussaoui (1,2), Pierre Dumouchel (1,2)

(1) Centre de Recherche Informatique de Montréal (CRIM), Quebec, Canada
(2) Ecole de Technologie Superieure (ETS), Montréal, Quebec, Canada

A novel approach to supervised dimensionality reduction is introduced, based on Gaussian Restricted Boltzmann Machines. The proposed model should be considered as the analogue of the probabilistic LDA, using undirected graphical models. The training algorithm of the model is presented while its close relation to the cosine distance is underlined. For the problem of speaker verification, we applied it to i-vectors and attained a significant improvement compared to the Fisher's Discriminant LDA projection using less than half of the number of eigenvectors required by LDA.

Index Terms: Speaker Recognition, Restricted Boltzmann Machines

Full Paper

Bibliographic reference.  Stafylakis, Themos / Kenny, Patrick / Senoussaoui, Mohammed / Dumouchel, Pierre (2012): "PLDA using Gaussian restricted boltzmann machines with application to speaker verification", In INTERSPEECH-2012, 1692-1695.