Nonlinear discriminant feature extraction for robust text-independent speaker recognition

Yochai Konig, Larry Heck, Mitch Weintraub, Kemal Sonmez

Research output: Contribution to conferencePaperpeer-review

3 Scopus citations

Abstract

We study a nonlinear discriminant analysis (NLDA) technique that extracts a speaker-discriminant feature set. Our approach is to train a multilayer perception (MLP) to maximize the separation between speakers by nonlinearly projecting a large set of acoustic features (e.g., several frames) to a lower-dimensional feature set. The extracted features are optimized to discriminate between speakers and to be robust to mismatched training and testing conditions. We train the MLP on a development set and apply it to the training and testing utterances. Our results show that by combining the NLDA-based system with a state of the art cepstrumbased system we improve the speaker verification performance on the 1997 NIST Speaker Recognition Evaluation set by 15% in average compared with our cepstrum-only system.

Original languageEnglish (US)
Pages72-75
Number of pages4
StatePublished - 2020
Externally publishedYes
EventWorkshop on Speaker Recognition and its Commercial and Forensic Applications, RLA2C 1998 - Avignon, France
Duration: Apr 20 1998Apr 23 1998

Conference

ConferenceWorkshop on Speaker Recognition and its Commercial and Forensic Applications, RLA2C 1998
CountryFrance
CityAvignon
Period4/20/984/23/98

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Signal Processing
  • Software

Fingerprint Dive into the research topics of 'Nonlinear discriminant feature extraction for robust text-independent speaker recognition'. Together they form a unique fingerprint.

Cite this