2 code implementations • 27 Oct 2022 • Anna Silnova, Niko Brümmer, Albert Swart, Lukáš Burget
It extends PSDA with the ability to model within and between-speaker variabilities in toroidal submanifolds of the hypersphere.
3 code implementations • 28 Mar 2022 • Niko Brümmer, Albert Swart, Ladislav Mošner, Anna Silnova, Oldřich Plchot, Themos Stafylakis, Lukáš Burget
In speaker recognition, where speech segments are mapped to embeddings on the unit hypersphere, two scoring backends are commonly used, namely cosine scoring or PLDA.
no code implementations • 15 Sep 2021 • Niko Brümmer
The Wishart distribution is the standard conjugate prior for the precision of the multivariate Gaussian likelihood, when the mean is known -- while the normal-Wishart can be used when the mean is also unknown.
no code implementations • 5 Sep 2021 • Josef Slavíček, Albert Swart, Michal Klčo, Niko Brümmer
We describe the Phonexia submission for the VoxCeleb Speaker Recognition Challenge 2021 (VoxSRC-21) in the unsupervised speaker verification track.
1 code implementation • 1 Apr 2021 • Niko Brümmer, Luciana Ferrer, Albert Swart
For perfect calibration, the Bayes error-rate is upper bounded by min(EER, P, 1-P), where EER is the equal-error-rate and P, 1-P are the prior probabilities of the competing hypotheses.
1 code implementation • 6 Apr 2020 • Anna Silnova, Niko Brümmer, Johan Rohdin, Themos Stafylakis, Lukáš Burget
We apply the proposed probabilistic embeddings as input to an agglomerative hierarchical clustering (AHC) algorithm to do diarization in the DIHARD'19 evaluation set.
no code implementations • 29 Sep 2017 • Niko Brümmer, Albert Swart
A standard recipe for spoken language recognition is to apply a Gaussian back-end to i-vectors.
no code implementations • 8 Mar 2016 • Niko Brümmer
This note compares two recently published machine learning methods for constructing flexible, but tractable families of variational hidden-variable posteriors.
no code implementations • 12 Oct 2015 • Niko Brümmer
We show here that the classical i-vector extractor recipe is actually a mean-field variational Bayes (VB) recipe.
no code implementations • 26 Mar 2014 • David A. van Leeuwen, Niko Brümmer
In this paper we study speaker linking (a. k. a.\ partitioning) given constraints of the distribution of speaker identities over speech recordings.
no code implementations • 24 Mar 2014 • Niko Brümmer, Albert Swart
We introduce a Bayesian solution for the problem in forensic speaker recognition, where there may be very little background material for estimating score calibration parameters.
no code implementations • 11 Feb 2014 • Niko Brümmer, Albert Swart, David van Leeuwen
In recent work on both generative and discriminative score to log-likelihood-ratio calibration, it was shown that linear transforms give good accuracy only for a limited range of operating points.
no code implementations • 24 Jan 2014 • Niko Brümmer
The Laplace approximation calls for the computation of second derivatives at the likelihood maximum.
no code implementations • 4 Nov 2013 • Niko Brümmer, Daniel Garcia-Romero
Score calibration enables automatic speaker recognizers to make cost-effective accept / reject decisions.
no code implementations • 30 Jul 2013 • Niko Brümmer, George Doddington
Prior-weighted logistic regression has become a standard tool for calibration in speaker recognition.
1 code implementation • 10 Apr 2013 • Niko Brümmer, Edward de Villiers
This poses the challenges of (i) how to decide what number of trials is enough, and (ii) how to process such large data sets with reasonable memory and CPU requirements.