Work Files Saved Searches
   My Account                                                  Search:   Quick/Number   Boolean   Advanced       Help   

 The Delphion Integrated View

  Buy Now:   Buy PDF- 23pp  PDF  |   File History  |   Other choices   
  Tools:  Citation Link  |  Add to Work File:    
  View:  Expand Details   |  INPADOC   |  Jump to: 
 Email this to a friend  Email this to a friend 
Title: US6269334: Nongaussian density estimation for the classification of acoustic feature vectors in speech recognition
[ Derwent Title ]

Country: US United States of America

View Images High


23 pages

Inventor: Basu, Sankar; Tenafly, NJ
Micchelli, Charles A.; Mogehan Lake, NY

Assignee: International Business Machines Corporation, Armonk, NY
other patents from INTERNATIONAL BUSINESS MACHINES CORPORATION (280070) (approx. 44,393)
 News, Profiles, Stocks and More about this company

Published / Filed: 2001-07-31 / 1998-06-25

Application Number: US1998000104553

IPC Code: Advanced: G10L 15/02;
Core: G10L 15/00;
IPC-7: G10L 13/00;

ECLA Code: G10L15/02;

U.S. Class: Current: 704/256.3; 704/255; 704/E15.004;
Original: 704/256; 704/255;

Field of Search: 904/231,255,240,256 704/231,255,256,240

Priority Number:
1998-06-25  US1998000104553

Abstract:     A statistical modeling paradigm for automatic machine recognition of speech uses mixtures of nongaussion statistical probability densities which provides improved recognition accuracy. Speech is modeled by building probability densities from functions of the form exp(-talpha/2) for t>=0 and alpha>0. Mixture components are constructed from different univariate functions. The mixture model is used in a maximum likelihood model of speech data.

Attorney, Agent or Firm: Whitham, Curtis & Whitham ; Kaufman, Esq., Stephen C. ;

Primary / Asst. Examiners: Zele, Krista; Opsasneck, Michael N.

INPADOC Legal Status: Show legal status actions

Family: None

First Claim:
Show all 8 claims
Having thus described our invention, what we claim as new and desire to secure by Letters Patent is as follows:     1. A computer implemented process for automatic machine recognition of speech comprising the steps of:
  • inputting acoustic data;
  • modeling input acoustic data using mixtures of nongaussian statistical probability densities constructed from a univariate function;
  • using a maximum likelihood model of speech data, iteratively generating values of mixture weights, means and variances until an acceptable density is found; and
  • storing a final density function form for decoding.

Background / Summary: Show background / summary

Drawing Descriptions: Show drawing descriptions

Description: Show description

Forward References: Show 6 U.S. patent(s) that reference this one

U.S. References: Go to Result Set: All U.S. references   |  Forward references (6)   |   Backward references (12)   |   Citation Link

Patent  Pub.Date  Inventor Assignee   Title
Get PDF - 33pp US4783804  1988-11 Juang et al.  American Telephone and Telegraph Company, AT&T Bell Laboratories Hidden Markov model speech recognition arrangement
Get PDF - 51pp US5148489  1992-09 Erell et al.  SRI International Method for spectral estimation to improve noise robustness for speech recognition
Get PDF - 14pp US5271088  1993-12 Bahler  ITT Corporation Automated sorting of voice messages through speaker spotting
Get PDF - 9pp US5473728  1995-12 Luginbuhl et al.  The United States of America as represented by the Secretary of the Navy Training of homoscedastic hidden Markov models for automatic speech recognition
Get PDF - 19pp US5694342  1997-12 Stein  The United States of America as represented by the Secretary of the Navy Method for detecting signals in non-Gaussian background clutter
Get PDF - 23pp US5706402  1998-01 Bell  The Salk Institute for Biological Studies Blind signal processing system employing information maximization to recover unknown signals through unsupervised minimization of output redundancy
Get PDF - 25pp US5737490  1998-04 Austin et al.  Apple Computer, Inc. Method and apparatus for constructing continuous parameter fenonic hidden markov models by replacing phonetic models with continous fenonic models
Get PDF - 25pp US5754681  1998-05 Watanabe et al.  ATR Interpreting Telecommunications Research Laboratories Signal pattern recognition apparatus comprising parameter training controller for training feature conversion parameters and discriminant functions
Get PDF - 9pp US5790758  1998-08 Streit  The United States of America as represented by the Secretary of the Navy Neural network architecture for gaussian components of a mixture density function
Get PDF - 25pp US5839105  1998-11 Ostendorf et al.  ATR Interpreting Telecommunications Research Laboratories Speaker-independent model generation apparatus and speech recognition apparatus each equipped with means for splitting state having maximum increase in likelihood
Get PDF - 15pp US5857169  1999-01 Seide  U.S. Philips Corporation Method and system for pattern recognition based on tree organized probability densities
Get PDF - 19pp US5864810  1999-01 Digalakis et al.  SRI International Method and apparatus for speech recognition adapted to an individual speaker
Foreign References: None

Other References:
  • Godsill et al, "Robust Noise Reduction For Speech and Audio Signals", IEEE, pp. 625-628, 1996.*
  • Laskey, "A Bayesian Approach to Clustering and Classification", IEEE pp. 179-183, 1991.*
  • Tugnait, "Parameter Identifiability of Multichannel ARMA Models of Linear Non-Gaussian Signals Via cumulant Matching", IEEE, IV 441-444, 1994.*
  • Frangoulis, "Vector Quantization of the Continuous Distributions of an HMM Speech Recogniser base on Mixtures of Continuous Distributions", IEEE, pp. 9-12, 1989.*
  • Pham et al, "Maximum Likelihood Estimation of a Class of Non-Gaussian Densities with Application to Lp Deconvolution", IEEE transactions on Acoustics, Speech, and Signal Processing, vol. 37, #1, Jan. 1989.*
  • Basu et al, "Maximum Likelihood Estimates for Exponential Type Density Families", Acoustics, Speech, and Signal Processing, Mar. 1999.*
  • Beadle et al, "Parameter Estimation for Non-Gaussian Autoregressive Processes", IEEE, pp. 3557-3560, 1997.*
  • Young et al, "The HTK Book", pp. 3-44, Entropic Cambridge Research Laboratory, Dec. 1997.*
  • Kuruoglu et al, "Nonlinear Autoregressive Modeling of Non-Gaussian Signals Using Lp Norm Techniques", IEEE, pp. 3533-3536, 1997.*
  • Zhuang et al, "Gaussian Mixture Density Modeling, Decomposition, and Applications", IEEE Transactions on Image Processing, vol. 5, #9, pp. 1293-1302, Sep. 1996. (10 pages) [ISI abstract]

  • Inquire Regarding Licensing

    Powered by Verity

    Plaques from Patent Awards      Gallery of Obscure PatentsNominate this for the Gallery...

    Thomson Reuters Copyright © 1997-2014 Thomson Reuters 
    Subscriptions  |  Web Seminars  |  Privacy  |  Terms & Conditions  |  Site Map  |  Contact Us  |  Help