Detail publikace

On the use of i-vector posterior distributions in Probabilistic Linear Discriminant Analysis

Originální název

On the use of i-vector posterior distributions in Probabilistic Linear Discriminant Analysis

Anglický název

On the use of i-vector posterior distributions in Probabilistic Linear Discriminant Analysis

Jazyk

en

Originální abstrakt

A PLDA model which exploits the uncertainty of the i-vector extraction process has been presented. We derived the formulation of the likelihood for a Gaussian PLDA model based on the i-vector posterior distribution, and illustrated a new PLDA model, where the inter-speaker variability is assumed to have an segment-dependent distribution, showing that we can rely on the standard PLDA framework simply replacing the likelihood definition.

Anglický abstrakt

A PLDA model which exploits the uncertainty of the i-vector extraction process has been presented. We derived the formulation of the likelihood for a Gaussian PLDA model based on the i-vector posterior distribution, and illustrated a new PLDA model, where the inter-speaker variability is assumed to have an segment-dependent distribution, showing that we can rely on the standard PLDA framework simply replacing the likelihood definition.

BibTex


@article{BUT111591,
  author="Sandro {Cumani} and Pietro {Laface} and Oldřich {Plchot}",
  title="On the use of i-vector posterior distributions in Probabilistic Linear Discriminant Analysis",
  annote="A PLDA model which exploits the uncertainty of the i-vector extraction process
has been presented. We derived the formulation of the likelihood for a Gaussian
PLDA model based on the i-vector posterior distribution, and illustrated a new
PLDA model, where the inter-speaker variability is assumed to have an
segment-dependent distribution, showing that we can rely on the standard PLDA
framework simply replacing the likelihood definition.",
  address="NEUVEDEN",
  chapter="111591",
  doi="10.1109/TASLP.2014.2308473",
  edition="NEUVEDEN",
  howpublished="print",
  institution="NEUVEDEN",
  number="4",
  volume="22",
  year="2014",
  month="march",
  pages="846--857",
  publisher="NEUVEDEN",
  type="journal article in Web of Science"
}