Conference paperCUMANI Sandro, PLCHOT Oldřich and LAFACE Pietro. Probabilistic Linear Discriminant Analysis Of IVector Posterior Distributions. In: Proceedings of ICASSP 2013. Vancouver: IEEE Signal Processing Society, 2013, pp. 76447648. ISBN 9781479903559.  Publication language:  english 

Original title:  Probabilistic Linear Discriminant Analysis Of IVector Posterior Distributions 

Title (cs):  Pravděpodobnostní lineární diskriminační analýza posteriorních rozložení ivektorů 

Pages:  76447648 

Proceedings:  Proceedings of ICASSP 2013 

Conference:  38th International Conference on Acoustics, Speech, and Signal Processing 

Place:  Vancouver, CA 

Year:  2013 

ISBN:  9781479903559 

Publisher:  IEEE Signal Processing Society 

URL:  http://www.fit.vutbr.cz/research/groups/speech/publi/2013/cumani_icassp2013_0007644.pdf [PDF] 

Keywords 

Speaker recognition, ivector, PLDA 
Annotation 

This article describes Probabilistic Linear Discriminant Analysis Of IVector Posterior Distributions. 
Abstract 

The ivector extraction process is affected by several factors such as
the noise level, the acoustic content of the observed features, and the
duration of the analyzed speech segment. These factors influence
both the ivector estimate and its uncertainty, represented by the i
vector posterior covariance. This paper present a new PLDA model
that, unlike the standard one, exploits the intrinsic ivector uncertainty.
Since short segments are known to decrease recognition accuracy,
and segment duration is the main factor affecting the ivector
covariance, we designed a set of experiments aiming at comparing
the standard and the new PLDA models on short speech cuts of variable
duration, randomly extracted from the conversations included in
the NIST SRE 2010 female telephone extended core condition. Our
results show that the new model outperforms the standard PLDA
when tested on short segments, and keeps the accuracy of the latter
for long enough utterances. In particular, the relative improvement
is up to 13% for the EER, 5% for DCF08, and 2.5% for DCF10. 
BibTeX: 

@INPROCEEDINGS{
author = {Sandro Cumani and Old{\v{r}}ich Plchot and Pietro
Laface},
title = {Probabilistic Linear Discriminant Analysis Of
IVector Posterior Distributions},
pages = {76447648},
booktitle = {Proceedings of ICASSP 2013},
year = 2013,
location = {Vancouver, CA},
publisher = {IEEE Signal Processing Society},
ISBN = {9781479903559},
language = {english},
url = {http://www.fit.vutbr.cz/research/view_pub.php.en?id=10312}
} 
