Patents by Inventor James H. Nealand

James H. Nealand has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Method and apparatus for training a text independent speaker recognition system using speech data with text labels
    Patent number: 7813927
    Abstract: There is provided an apparatus for providing a Text Independent (TI) speaker recognition mode in a Text Dependent (TD) Hidden Markov Model (HMM) speaker recognition system and/or a Text Constrained (TC) HMM speaker recognition system. The apparatus includes a Gaussian Mixture Model (GMM) generator and a Gaussian weight normalizer. The GMM generator is for creating a GMM by pooling Gaussians from a plurality of HMM states. The Gaussian weight normalizer is for normalizing Gaussian weights with respect to the plurality of HMM states.
    Type: Grant
    Filed: June 4, 2008
    Date of Patent: October 12, 2010
    Assignee: Nuance Communications, Inc.
    Inventors: Jiri Navratil, James H. Nealand, Jason W. Pelecanos, Ganesh N. Ramaswamy, Ran D. Zilca
  • Method and apparatus for training a text independent speaker recognition system using speech data with text labels
    Patent number: 7447633
    Abstract: There is provided an apparatus for providing a Text Independent (TI) speaker recognition mode in a Text Dependent (TD) Hidden Markov Model (HMM) speaker recognition system and/or a Text Constrained (TC) HMM speaker recognition system. The apparatus includes a Gaussian Mixture Model (GMM) generator and a Gaussian weight normalizer. The GMM generator is for creating a GMM by pooling Gaussians from a plurality of HMM states. The Gaussian weight normalizer is for normalizing Gaussian weights with respect to the plurality of HMM states.
    Type: Grant
    Filed: November 22, 2004
    Date of Patent: November 4, 2008
    Assignee: International Business Machines Corporation
    Inventors: Jiri Navratil, James H. Nealand, Jason W. Pelecanos, Ganesh N. Ramaswamy, Ran D. Zilca
  • METHOD AND APPARATUS FOR TRAINING A TEXT INDEPENDENT SPEAKER RECOGNITION SYSTEM USING SPEECH DATA WITH TEXT LABELS
    Publication number: 20080235020
    Abstract: There is provided an apparatus for providing a Text Independent (TI) speaker recognition mode in a Text Dependent (TD) Hidden Markov Model (HMM) speaker recognition system and/or a Text Constrained (TC) HMM speaker recognition system. The apparatus includes a Gaussian Mixture Model (GMM) generator and a Gaussian weight normalizer. The GMM generator is for creating a GMM by pooling Gaussians from a plurality of HMM states. The Gaussian weight normalizer is for normalizing Gaussian weights with respect to the plurality of HMM states.
    Type: Application
    Filed: June 4, 2008
    Publication date: September 25, 2008
    Inventors: Jiri Navratil, James H. Nealand, Jason W. Pelecanos, Ganesh N. Ramaswamy, Ran D. Zilca