Publication Details
Discriminatively Trained Probabilistic Linear Discriminant Analysis for Speaker Verification
Plchot Oldřich, Ing., Ph.D. (DCGM FIT BUT)
Cumani Sandro (POLITO)
Glembek Ondřej, Ing., Ph.D. (DCGM FIT BUT)
Matějka Pavel, Ing., Ph.D. (DCGM FIT BUT)
Brümmer Niko (Agnitio)
Speaker verification, Discriminative training, Probabilistic Linear Discriminant Analysis
This paper is on Discriminatively Trained Probabilistic Linear Discriminant Analysis (PLDA) for Speaker Verification.
Recently, i-vector extraction and Probabilistic Linear Discriminant Analysis (PLDA) have proven to provide state-of-the-art speaker verification performance. In this paper, the speaker verification score for a pair of i-vectors representing a trial is computed with a functional form derived from the successful PLDA generative model. In our case, however, parameters of this function are estimated based on a discriminative training criterion. We propose to use the objective function to directly address the task in speaker verification: discrimination between same-speaker and different-speaker trials. Compared with a baseline which uses a generatively trained PLDA model, discriminative training provides up to 40% relative improvement on the NIST SRE 2010 evaluation task.
@INPROCEEDINGS{FITPUB9653,
author = "Luk\'{a}\v{s} Burget and Old\v{r}ich Plchot and Sandro Cumani and Ond\v{r}ej Glembek and Pavel Mat\v{e}jka and Niko Br{\"{u}}mmer",
title = "Discriminatively Trained Probabilistic Linear Discriminant Analysis for Speaker Verification",
pages = "4832--4835",
booktitle = "Proceedings of the 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011",
year = 2011,
location = "Praha, CZ",
publisher = "IEEE Signal Processing Society",
ISBN = "978-1-4577-0537-3",
doi = "10.1109/ICASSP.2011.5947437",
language = "english",
url = "https://www.fit.vut.cz/research/publication/9653"
}