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


 The Delphion Integrated View

  Buy Now:   Buy PDF- 10pp  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: US5970239: Apparatus and method for performing model estimation utilizing a discriminant measure
[ Derwent Title ]


Country: US United States of America

View Images High
Resolution

 Low
 Resolution

 
10 pages

 
Inventor: Bahl, Lalit Rai; Amawalk, NY
Padmanabhan, Mukund; Ossining, 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: 1999-10-19 / 1997-08-11

Application Number: US1997000908120

IPC Code: Advanced: G06F 9/455;
IPC-7: G06F 9/455;

U.S. Class: Current: 704/245; 704/231; 704/236;
Original: 395/500.23; 704/231; 704/236;

Field of Search: 364/578 704/231,236

Priority Number:
1997-08-11  US1997000908120

Abstract:     Method for performing acoustic model estimation to optimize classification accuracy on speaker derived feature vectors with respect to a plurality of classes corresponding to phones to which a plurality of acoustic models respectively correspond comprises: (a) initializing an acoustic model for each phone; (b) evaluating the merit of the acoustic model initialized for each phone utilizing an objective function having a two component discriminant measure capable of characterizing each phone whereby a first component is defined as a probability that the model for the phone assigns to feature vectors from the phone and a second component is defined as a probability that the model for the phone assigns to feature vectors from other phones; (c) adapting the model for selected phones so as to increase the first component for the phone or decrease the second component for the phone, the adapting step yielding a new model for each selected phone; (d) evaluating the merit of the new models for each phone adapted in step (c) utilizing the two component measure; (e) comparing results of the evaluation of step (b) with results of the evaluation of step (d) for each phone, and if the first component has increased or the second component has decreased, the new model is kept for that phone, else the model originally initialized is kept; (f) estimating parameters associated with each model kept for each phone in order to optimize the function; and (g) evaluating termination criterion to determine if the parameters of the models are optimized.

Attorney, Agent or Firm: F. Chau & Associates, LLP ;

Primary / Asst. Examiners: Teska, Kevin J.; Broda, Samuel

Maintenance Status: E1 Expired  Check current status

INPADOC Legal Status: Show legal status actions

Family: None

First Claim:
Show all 23 claims
What is claimed is:     1. Apparatus for performing acoustic model estimation in order to optimize classification accuracy on feature vectors derived from a speaker with respect to a plurality of classes corresponding to phones to which a plurality of acoustic models respectively correspond, the apparatus comprising:
  • means for initializing an acoustic model for each class;
  • first means for evaluating the merit of the acoustic model initialized for each phone utilizing an objective function having a two component discriminant measure capable of characterizing each phone whereby a first component is defined as a probability that the acoustic model for the phone assigns to the feature vectors from the phone and a second component is defined as a probability that the acoustic model for the phone assigns to the feature vectors from other phones;
  • means for adapting the acoustic model for selected phones so as to one of increase the first component of the discriminant measure for the phone and decrease the second component of the discriminant measure for the phone, the adapting means yielding a new acoustic model for each selected phone;
  • second means for evaluating the merit of the new acoustic models for each phone adapted by the adapting means utilizing the two component discriminant measure;
  • means for comparing results obtained by the first evaluating means with results obtained by the second evaluating means for each phone, and if one of the first component of the discriminant measure has increased and the second component of the discriminant measure has decreased, then the new acoustic model is kept for that phone, else the acoustic model originally initialized is kept;
  • means for estimating parameters associated with each acoustic model kept for each phone in order to substantially optimize the objective function; and
  • third means for evaluating termination criterion to determine if the parameters of the acoustic models are substantially optimized.


Background / Summary: Show background / summary

Drawing Descriptions: Show drawing descriptions

Description: Show description

Forward References: Show 10 U.S. patent(s) that reference this one

       
U.S. References: Go to Result Set: All U.S. references   |  Forward references (10)   |   Backward references (6)   |   Citation Link

Buy
PDF
Patent  Pub.Date  Inventor Assignee   Title
Get PDF - 15pp US5195167  1993-03 Bahl et al.  International Business Machines Corporation Apparatus and method of grouping utterances of a phoneme into context-dependent categories based on sound-similarity for automatic speech recognition
Get PDF - 12pp US5222146  1993-06 Bahl et al.  International Business Machines Corporation Speech recognition apparatus having a speech coder outputting acoustic prototype ranks
Get PDF - 15pp US5455889  1995-10 Bahl et al.  International Business Machines Corporation Labelling speech using context-dependent acoustic prototypes
Get PDF - 14pp US5497447  1996-03 Bahl et al.  International Business Machines Corporation Speech coding apparatus having acoustic prototype vectors generated by tying to elementary models and clustering around reference vectors
Get PDF - 15pp US5615299  1997-03 Bahl et al.  International Business Machines Corporation Speech recognition using dynamic features
Get PDF - 12pp US5787394  1998-07 Bahl et al.  International Business Machines Corporation State-dependent speaker clustering for speaker adaptation
       
Foreign References: None

Other References:
  • L.R. Bahl, P.F. Brown, P.V. deSouza, R.L. Mercer in "Maximum Mutual Information Estimation of Hidden Markov Model Parameters for Speech Recognition", Proceedings of the ICASSP, pp. 49-52, 1986.
  • B.H. Juang, W. Chou, C.H. Lee in "Minimum Classification Error Rate Methods for Speech Recognition", IEEE Trans. on Speech and Audio Processing, vol. 5, pp. 257-265, May 1997. (9 pages) Cited by 4 patents [ISI abstract]
  • A.P. Dempster, N.M. Laird, D.B. Rubin in "Maximum Likelihood Estimation from Incomplete Data via the EM Algorithm", Journal of the Royal Statistical Society (B), vol. 39, No. 1, pp. 1-38, 1979.
  • R.O. Duda and P.E. Hart in "Pattern Classification and Scene Analysis", Wiley, New York, 1973.
  • R. Lippman in "Pattern Classification Using Neural Networks", IEEE Communications Magazine, pp. 11:47-64, 1989.
  • Y. Normandin in "Optimal Splitting of HMM Gaussian Mixture Components with MMIE Training", Proceedings of the ICASSP, pp. 449-452, 1995.
  • A.J. Viterbi in "Error Bounds for Convolutional Codes and An Asymptotically Optimum Decoding Algorithm", IEEE Trans. on Information Theory, vol. IT-13, pp. 260-269, Apr. 1967.


  • 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