Loading...

Localized discriminative Gaussian process latent variable model for text-dependent speaker verification

Maghsoodi, N ; Sharif University of Technology | 2016

572 Viewed
  1. Type of Document: Article
  2. Publisher: i6doc.com publication , 2016
  3. Abstract:
  4. The duration of utterances is one of the effective factors on the performance of speaker verification systems. Text dependent speaker verification suffers from both short duration and unmatched content between enrollment and test segments. In this paper, we use Discriminative Gaussian Process Latent Variable Model (DGPLVM) to deal with the uncertainty caused by short duration. This is the first attempt to utilize Gaussian Process for speaker verification. Also, to manage the unmatched content between enrollment and test segments we proposed the localized-DGPLVM that trains DGPLVM for each phrase in dataset. Experiments show the relative improvement of 27.4% in EER on RSR2015
  5. Keywords:
  6. Artificial intelligence ; Gaussian distribution ; Gaussian noise (electronic) ; Learning systems ; Neural networks ; Principal component analysis ; Statistical tests ; Gaussian Processes ; Latent variable modeling ; Short durations ; Speaker verification ; Speaker verification system ; Speech recognition
  7. Source: 24th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2016, 27 April 2016 through 29 April 2016 ; 2016 , Pages 183-188 ; 9782875870278 (ISBN)