# Pairwise Discriminative Neural PLDA for Speaker Verification

arXiv: https://arxiv.org/abs/2001.07034 · HTML: https://mlsanity.com/paper/2001.07034/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jan 2020.

## Built on by (most cited first, 1 of 1)

- [I4U System Description for NIST SRE'20 CTS Challenge](https://mlsanity.com/paper/2211.01091/) (2022)

## Abstract (arXiv)

The state-of-art approach to speaker verification involves the extraction of discriminative embeddings like x-vectors followed by a generative model back-end using a probabilistic linear discriminant analysis (PLDA). In this paper, we propose a Pairwise neural discriminative model for the task of speaker verification which operates on a pair of speaker embeddings such as x-vectors/i-vectors and outputs a score that can be considered as a scaled log-likelihood ratio. We construct a differentiable cost function which approximates speaker verification loss, namely the minimum detection cost. The pre-processing steps of linear discriminant analysis (LDA), unit length normalization and within class covariance normalization are all modeled as layers of a neural model and the speaker verification cost functions can be back-propagated through these layers during training. We also explore regularization techniques to prevent overfitting, which is a major concern in using discriminative back-end models for verification tasks. The experiments are performed on the NIST SRE 2018 development and evaluation datasets. We observe average relative improvements of 8% in CMN2 condition and 30% in VAST condition over the PLDA baseline system.