Publication detail

Toroidal Probabilistic Spherical Discriminant Analysis

SILNOVA, A. BRUMMER, J. SWART, A. BURGET, L.

Original Title

Toroidal Probabilistic Spherical Discriminant Analysis

Type

conference paper

Language

English

Original Abstract

n speaker recognition, where speech segments are mapped to embeddings on the unit hypersphere, two scoring back-ends are commonly used, namely cosine scoring and PLDA. We have recently proposed PSDA, an analog to PLDA that uses Von Mises-Fisher distributions instead of Gaussians. In this paper, we present toroidal PSDA (T-PSDA). It extends PSDA with the ability to model within and between-speaker variabilities in toroidal submanifolds of the hypersphere. Like PLDA and PSDA, the model allows closed-form scoring and closed-form EM updates for training. On VoxCeleb, we find T-PSDA accu- racy on par with cosine scoring, while PLDA accuracy is infe- rior. On NIST SRE'21 we find that T-PSDA gives large accu- racy gains compared to both cosine scoring and PLDA.

Keywords

speaker recognition, PSDA, Von Mises-Fishe

Authors

SILNOVA, A.; BRUMMER, J.; SWART, A.; BURGET, L.

Released

4. 6. 2023

Publisher

IEEE Signal Processing Society

Location

Rhodes Island

ISBN

978-1-7281-6327-7

Book

Proceedings of ICASSP 2023

Pages from

1

Pages to

5

Pages count

5

URL

BibTex

@inproceedings{BUT185199,
  author="Anna {Silnova} and Johan Nikolaas Langenhoven {Brummer} and Albert du Preez {Swart} and Lukáš {Burget}",
  title="Toroidal Probabilistic Spherical Discriminant Analysis",
  booktitle="Proceedings of ICASSP 2023",
  year="2023",
  pages="1--5",
  publisher="IEEE Signal Processing Society",
  address="Rhodes Island",
  doi="10.1109/ICASSP49357.2023.10095580",
  isbn="978-1-7281-6327-7",
  url="https://ieeexplore.ieee.org/document/10095580"
}