Comparison of scoring methods used in speaker recognition with Joint Factor Analysis

Ondrej Glembek, Lukás Burget, Najim Dehak, Niko Brümmer, Patrick KennyICASSP 2009 · OpenReview

Abstract

The aim of this paper is to compare different log-likelihood scoring methods, that different sites used in the latest state-of-the-art Joint Factor Analysis (JFA) Speaker Recognition systems. The algorithms use various assumptions and have been derived from various approximations of the objective functions of JFA. We compare the techniques in terms of speed and performance. We show, that approximations of the true log-likelihood ratio (LLR) may lead to significant speedup without any loss in performance.
BibTeX
@inproceedings{glembek2009comparison,
  author = {Ondrej Glembek and Lukás Burget and Najim Dehak and Niko Brümmer and Patrick Kenny},
  title = {{Comparison of scoring methods used in speaker recognition with Joint Factor Analysis}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2009},
  url = {https://openreview.net/forum?id=HAeyyD38kan},
}

Loading related content...

Figure