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

 The Delphion Integrated View

  Buy Now:   Buy PDF- 28pp  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: US5072452: Automatic determination of labels and Markov word models in a speech recognition system
[ Derwent Title ]

Country: US United States of America

View Images High


28 pages

Inventor: Brown, Peter F.; New York, NY
De Souza, Peter V.; Yorktown Heights, NY
Nahomoo, David; White Plains, NY
Picheny, Michael A.; White Plains, 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: 1991-12-10 / 1989-11-02

Application Number: US1989000431720

IPC Code: Advanced: G10L 15/14;
IPC-7: G10L 5/06;

ECLA Code: G10L15/14; T05K999/99;

U.S. Class: Current: 704/256.4;
Original: 381/043;

Field of Search: 381/041-46 364/513,513.5 382/014-15

Priority Number:
1989-11-02  US1989000431720
1987-10-30  US1987000115505

Abstract:     In a Markov model speech recognition system, an acoustic processor generates one label after another selected from an alphabet of labels. Each vocabulary word is represented as a baseform constructed of a sequence of Markov models. Each Markov model is stored in a computer memory as (a) a plurality of states; (b) a plurality of arcs, each extending from a state to a state with a respective stored probability; and (c) stored label output probabilities, each indicating the likelihood of a given label being produced at a certain arc. Word likelihood based on acoustic characteristics is determined by matching a string of labels generated by the acoustic processor against the probabilities stored for each word baseform. Improved models of words are obtained by specifying label parameters and constructing word baseforms interdependently and iteratively.

Attorney, Agent or Firm: Schechter, Marc D. ;

Primary / Asst. Examiners: Harkcom, Gary V.; Merecki, John

Maintenance Status: E2 Expired  Check current status

INPADOC Legal Status: Show legal status actions          Buy Now: Family Legal Status Report

Related Applications:
Application Number Filed Patent Pub. Date  Title
US1987000115505 1987-10-30       

Parent Case:     This is a continuation of application Ser. No. 115,505, filed Oct. 30, 1987, now abandoned.

Designated Country: DE FR GB 

Family: Show 9 known family members

First Claim:
Show all 21 claims
We claim:     1. Speech processing apparatus comprising:
  • an acoustic processor for producing as a first output, in response to speech input, one label after another at successive time intervals, each label being selected from an alphabet of labels, each label having parameter values;
  • dictionary means for storing statistical data for each of a plurality of vocabulary words as Markov model word baseforms, wherein each baseform is characterized by a sequence of Markov models, at least one word baseform containing at least one Markov model at different locations in the sequence, each Markov model having a plurality of arcs, wherein the dictionary means includes storage for (i) the respective probability of each arc in each Markov model, and (ii) a respective probability of producing each label in the alphabet at each of some arcs in each Markov model
  • means, coupled to said acoustic processor, for re-specifying the parameter values of the labels in the alphabet which can be produced as outputs of the acoustic processor; and
  • baseform constructor means, coupled to said dictionary means, for up-dating the stored data for the Markov model word baseforms from labels generated by the acoustic processor based on the re-specified parameter values;
  • wherein said label re-specifying means re-specifies the parameter values of labels based on the up-dated stored data for the Markov model word baseforms;
  • wherein said acoustic processor produces as a second output one feature vector after another at the successive time intervals;
  • wherein each different Markov model corresponds to one respective label; and wherein said label re-specifying means includes:
    • alignment processor means for aligning a string of labels generated by the acoustic processor against a word baseform stored in the dictionary means, said alignment processor means aligning successive substrings in the string with successive Markov models in the word baseform; and
    • estimator means for receiving as input from the acoustic processor the feature vectors corresponding to the labels aligned with a given Markov model and computing means and covariance values of the feature vectors received for the given Markov model; and
    • label specifier means, coupled to the estimator means, for storing (i) the mean and covariance values of the feature vectors corresponding to the labels aligned with each Markov model, as (ii) the parameter values of the label corresponding to the Markov model.

Background / Summary: Show background / summary

Drawing Descriptions: Show drawing descriptions

Description: Show description

Forward References: Show 65 U.S. patent(s) that reference this one

U.S. References: Go to Result Set: All U.S. references   |  Forward references (65)   |   Backward references (19)   |   Citation Link

Patent  Pub.Date  Inventor Assignee   Title
Get PDF - 138pp USRE26772  1970-01 Lazarus   INTERPRETATION MACHINE
Get PDF - 14pp US3333248* 1967-07 Greenberg et al.    
Get PDF - 176pp US3440617  1969-04 Lesti   SIGNAL RESPONSIVE SYSTEMS
Get PDF - 36pp US3950733  1976-04 Cooper et al.  Nestor Associates Information processing system
Get PDF - 25pp US4177448  1979-12 Brayton  International Business Machines Corporation Character recognition system and method multi-bit curve vector processing
Get PDF - 44pp US4181821  1980-01 Pirz et al.  Bell Telephone Laboratories, Incorporated Multiple template speech recognition system
Get PDF - 18pp US4309756  1982-01 Beckler   Method of automatically evaluating source language logic condition sets and of compiling machine executable instructions directly therefrom
Get PDF - 22pp US4327354  1982-04 Persoon  U.S. Philips Corporation Learning device for digital signal pattern recognition
Get PDF - 48pp US4336421  1982-06 Welch et al.  Threshold Technology, Inc. Apparatus and method for recognizing spoken words
Get PDF - 28pp US4412098  1983-10 An  Interstate Electronics Corporation Audio signal recognition computer
Get PDF - 9pp US4451929  1984-05 Yoshida  Hajime Industries Ltd. Pattern discrimination method
Get PDF - 7pp US4618984  1986-10 Das et al.  International Business Machines Corporation Adaptive automatic discrete utterance recognition
Get PDF - 13pp US4741036  1988-04 Bahl et al.  International Business Machines Corporation Determination of phone weights for markov models in a speech recognition system
Get PDF - 54pp US4748670  1988-05 Bahl et al.  International Business Machines Corporation Apparatus and method for determining a likely word sequence from labels generated by an acoustic processor
Get PDF - 12pp US4751737  1988-06 Gerson et al.  Motorola Inc. Template generation method in a speech recognition system
Get PDF - 35pp US4759068  1988-07 Bahl et al.  International Business Machines Corporation Constructing Markov models of words from multiple utterances
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 - 32pp US4827521  1989-05 Bahl et al.  International Business Machines Corporation Training of markov models used in a speech recognition system
  * some details unavailable
Foreign References:
Publication Date IPC Code Assignee   Title
Get PDF - 61pp EP0238691 1987-09  G10L 5/06 IBM Method and apparatus for determining a likely word sequence from labels generated by an acoustic processor 
Get PDF - 39pp EP0238697 1987-09  G10L 5/06 IBM Method of constructing baseform models of words from multiple utterances for speech recognition 
Get PDF - 18pp EP0243009 1987-10  G10L 5/06 IBM Speech recognition system using Markov models 

Other Abstract Info: DERABS G89-138957

Other References:
  • Nadas, A., et al., "Continuous Speech Recognition with Automatically Selected Acoustic Prototypes Obtained by Either Bootstrapping or Clustering," IEEE CH1610-5/81, pp. 1153-1155.

  • 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