INTERSPEECH 2011
12th Annual Conference of the International Speech Communication Association

Florence, Italy
August 27-31. 2011

Mixture of PLDA Models in i-vector Space for Gender-Independent Speaker Recognition

Mohammed Senoussaoui (1), Patrick Kenny (1), Niko Brümmer (2), Edward de Villiers (2), Pierre Dumouchel (1)

(1) CRIM, Canada
(2) Agnitio, South Africa

The Speaker Recognition community that participates in NIST evaluations has concentrated on designing gender- and channelconditioned systems. In the real word, this conditioning is not feasible. Our main purpose in this work is to propose a mixture of Probabilistic Linear Discriminant Analysis models (PLDA) as a solution for making systems independent of speaker gender. In order to show the effectiveness of the mixture model, we first experiment on 2010 NIST telephone speech (det5), where we prove that there is no loss of accuracy compared with a baseline gender-dependent model. We also test with success the mixture model on a more realistic situation where there are cross-gender trials. Furthermore, we report results on microphone speech for the det1, det2, det3 and det4 tasks to confirm the effectiveness of the mixture model.

Full Paper

Bibliographic reference.  Senoussaoui, Mohammed / Kenny, Patrick / Brümmer, Niko / Villiers, Edward de / Dumouchel, Pierre (2011): "Mixture of PLDA models in i-vector space for gender-independent speaker recognition", In INTERSPEECH-2011, 25-28.