eduzhai > Applied Sciences > Engineering >

A Generalized Framework for Domain Adaptation of PLDA in Speaker Recognition

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 5 pages

Abstract: This paper proposes a generalized framework for domain adaptation ofProbabilistic Linear Discriminant Analysis (PLDA) in speaker recognition. Itnot only includes several existing supervised and unsupervised domainadaptation methods but also makes possible more flexible usage of availabledata in different domains. In particular, we introduce here the two newtechniques described below. (1) Correlation-alignment-based interpolation and(2) covariance regularization. The proposed correlation-alignment-basedinterpolation method decreases minCprimary up to 30.5 as compared with thatfrom an out-of-domain PLDA model before adaptation, and minCprimary is also5.5 lower than with a conventional linear interpolation method with optimalinterpolation weights. Further, the proposed regularization technique ensuresrobustness in interpolations w.r.t. varying interpolation weights, which inpractice is essential.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×