DeepAI AI Chat
Log In Sign Up

Investigation of Different Calibration Methods for Deep Speaker Embedding based Verification Systems

by   Galina Lavrentyeva, et al.
Speech Technology Center

Deep speaker embedding extractors have already become new state-of-the-art systems in the speaker verification field. However, the problem of verification score calibration for such systems often remains out of focus. An irrelevant score calibration leads to serious issues, especially in the case of unknown acoustic conditions, even if we use a strong speaker verification system in terms of threshold-free metrics. This paper presents an investigation over several methods of score calibration: a classical approach based on the logistic regression model; the recently presented magnitude estimation network MagnetO that uses activations from the pooling layer of the trained deep speaker extractor and generalization of such approach based on separate scale and offset prediction neural networks. An additional focus of this research is to estimate the impact of score normalization on the calibration performance of the system. The obtained results demonstrate that there are no serious problems if in-domain development data are used for calibration tuning. Otherwise, a trade-off between good calibration performance and threshold-free system quality arises. In most cases using adaptive s-norm helps to stabilize score distributions and to improve system performance. Meanwhile, some experiments demonstrate that novel approaches have their limits in score stabilization on several datasets.


page 1

page 2

page 3

page 4


A Speaker Verification Backend with Robust Performance across Conditions

In this paper, we address the problem of speaker verification in conditi...

A discriminative condition-aware backend for speaker verification

We present a scoring approach for speaker verification that mimics the s...

A Speaker Verification Backend for Improved Calibration Performance across Varying Conditions

In a recent work, we presented a discriminative backend for speaker veri...

Deep Normalization for Speaker Vectors

Deep speaker embedding has demonstrated state-of-the-art performance in ...

Robust Speaker Recognition with Transformers Using wav2vec 2.0

Recent advances in unsupervised speech representation learning discover ...

Classifier Calibration: with implications to threat scores in cybersecurity

This paper explores the calibration of a classifier output score in bina...