US Patent No: 5,075,896

Number of patents in Portfolio can not be more than 2000

Character and phoneme recognition based on probability clustering

Stats

ATTORNEY / AGENT: (SPONSORED)
 

Importance

Loading Importance Indicators... loading....

Abstract

Prior to character or phoneme recognition, a classifier provides a respective probability list for each of a sequence of sample characters or phonemes, each probability list indicating the respective sample's probability for each character or phoneme type. These probability lists are clustered in character or phoneme probability space, in which each dimension corresponds to the probability that a character or phoneme candidate is an instance of a specific character or phoneme type. For each resulting cluster, data is stored indicating its cluster ID and a probability list indicating the probability of each type at the cluster's center. Then, during recognition, a probability cluster identifier compares the probability list for each candidate with the probability list for each cluster to find the nearest cluster. The cluster identifier then provides the nearest cluster's cluster ID to a constraint satisfier that attempts to recognize the candidate based on rules, patterns, or a combination of rules and patterns. If necessary, the constraint satisfier uses the cluster ID to retrieve the stored probability list of the cluster to assist it in recognition.

Loading the Abstract Image... loading....

First Claim

Related Publications

Loading Related Publications... loading....

Patent Owner(s)

Patent OwnerAddressTotal Patents
XEROX CORPORATIONSTAMFORD, CT17085

International Classification(s)

  • [Classification Symbol]
  • [Patents Count]

Inventor(s)

Inventor Name Address # of filed Patents Total Citations
Spitz, A Lawrence Palo Alto, CA 10 367
Wilcox, Lynn D Palo Alto, CA 58 1560

Cited Art

Patent Info (Count) # Cites Year
 
DRAGON SYSTEMS, INC. (2)
4,837,831 Method for creating and using multiple-word sound models in speech recognition 77 1986
4,903,305 Method for representing word models for use in speech recognition 99 1989
 
BELL TELEPHONE LABORATORIES, INCORPORATED (1)
4,783,804 Hidden Markov model speech recognition arrangement 87 1985
 
ELECTRO-SENSORS, INC. (1)
4,541,115 Pattern processing system 38 1983
 
NESTOR, INC. (1)
4,958,375 Parallel, multi-unit, adaptive pattern classification system using inter-unit correlations and an intra-unit class separator methodology 54 1988
 
NUANCE COMMUNICATIONS, INC. (1)
4,773,099 Pattern classification means for use in a pattern recognition system 51 1985

Patent Citation Ranking

Forward Cites

Patent Info (Count) # Cites Year
 
CANON KABUSHIKI KAISHA (13)
5,982,933 Information processing method, information processing apparatus, and storage medium 13 1996
6,567,552 Image processing method and apparatus 1 1997
7,310,600 Language recognition using a similarity measure 11 2000
7,212,968 Pattern matching method and apparatus 8 2000
6,882,970 Language recognition using sequence frequency 17 2000
7,054,812 Database annotation and retrieval 28 2001
6,873,993 Indexing method and apparatus 28 2001
6,990,448 Database annotation and retrieval including phoneme data 18 2001
7,240,003 Database annotation and retrieval 4 2001
7,337,116 Speech processing system 20 2001
6,801,891 Speech processing system 7 2001
7,257,533 Database searching and retrieval using phoneme and word lattice 5 2005
7,295,980 Pattern matching method and apparatus 5 2006
 
INTERNATIONAL BUSINESS MACHINES CORPORATION (9)
5,343,537 Statistical mixture approach to automatic handwriting recognition 31 1991
5,544,257 Continuous parameter hidden Markov model approach to automatic handwriting recognition 20 1992
5,335,289 Recognition of characters in cursive script 14 1992
5,644,652 System and method for automatic handwriting recognition with a writer-independent chirographic label alphabet 50 1995
5,636,291 Continuous parameter hidden Markov model approach to automatic handwriting recognition 13 1995
6,067,514 Method for automatically punctuating a speech utterance in a continuous speech recognition system 21 1998
7,343,041 Handwritten word recognition using nearest neighbor techniques that allow adaptive learning 2 2002
7,466,861 Method for outputting character recognition results 2 2005
7,697,760 Handwritten word recognition using nearest neighbor techniques that allow adaptive learning 6 2008
 
GOOGLE INC. (5)
5,745,649 Automated speech recognition using a plurality of different multilayer perception structures to model a plurality of distinct phoneme categories 12 1997
8,175,394 Shape clustering in post optical character recognition processing 0 2006
8,111,927 Shape clustering in post optical character recognition processing 1 2010
8,131,085 Shape clustering in post optical character recognition processing 0 2011
8,170,351 Shape clustering in post optical character recognition processing 0 2011
 
XEROX CORPORATION (5)
5,442,778 Scatter-gather: a cluster-based method and apparatus for browsing large document collections 149 1991
5,483,650 Method of constant interaction-time clustering applied to document browsing 27 1993
5,592,568 Word spotting in bitmap images using context-sensitive character models without baselines 15 1995
5,778,095 Classification of scanned symbols into equivalence classes 12 1995
5,787,422 Method and apparatus for information accesss employing overlapping clusters 75 1996
 
SRI INTERNATIONAL (4)
5,864,810 Method and apparatus for speech recognition adapted to an individual speaker 74 1995
5,634,086 Method and apparatus for voice-interactive language instruction 83 1995
6,055,498 Method and apparatus for automatic text-independent grading of pronunciation for language instruction 48 1997
6,226,611 Method and system for automatic text-independent grading of pronunciation for language instruction 25 2000
 
HITACHI, LTD. (3)
5,634,134 Method and apparatus for determining character and character mode for multi-lingual keyboard based on input characters 45 1992
5,329,596 Automatic clustering method 28 1992
5,526,259 Method and apparatus for inputting text 25 1994
 
LUCENT TECHNOLOGIES INC. (3)
6,173,262 Text-to-speech system with automatically trained phrasing rules 11 1995
5,754,695 Degraded gray-scale document recognition using pseudo two-dimensional hidden Markov models and N-best hypotheses 25 1996
6,003,005 Text-to-speech system and a method and apparatus for training the same based upon intonational feature annotations of input text 13 1997
 
MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. (3)
5,909,508 Parallel image-clustering apparatus 4 1996
5,806,030 Low complexity, high accuracy clustering method for speech recognizer 29 1996
6,662,180 Method for searching in large databases of automatically recognized text 44 1999
 
APPLE INC. (2)
5,535,305 Sub-partitioned vector quantization of probability density functions 35 1992
5,828,999 Method and system for deriving a large-span semantic language model for large-vocabulary recognition systems 34 1996
 
FACEBOOK, INC. (2)
7,366,352 Method and apparatus for performing fast closest match in pattern recognition 1 2003
7,724,963 Apparatus for performing fast closest match in pattern recognition 1 2008
 
MASSACHUSETTS INSTITUTE OF TECHNOLOGY (2)
5,537,488 Pattern recognition system with statistical classification 32 1993
5,703,964 Pattern recognition system with statistical classification 47 1996
 
MOTOROLA MOBILITY LLC (2)
5,854,855 Method and system using meta-classes and polynomial discriminant functions for handwriting recognition 33 1995
5,802,205 Method and system for lexical processing 22 1995
 
AVAYA INC. (1)
5,982,926 Real-time image enhancement techniques 55 1995
 
BABYLON LTD. (1)
6,298,158 Recognition and translation system and method 0 1997
 
BIOSIS (1)
7,139,755 Method and apparatus for providing comprehensive search results in response to user queries entered over a computer network 12 2001
 
DIALOG CORPORATION PLC, THE (1)
6,137,911 Test classification system and method 156 1997
 
EASTMAN KODAK COMPANY (1)
5,325,445 Feature classification using supervised statistical pattern recognition 59 1992
 
FUJI XEROX CO., LTD. (1)
5,943,443 Method and apparatus for image based document processing 46 1997
 
GENERAL ELECTRIC COMPANY (1)
5,742,522 Adaptive, on line, statistical method and apparatus for detection of broken bars in motors by passive motor current monitoring and digital torque estimation 23 1996
 
GRUMMAN AEROSPACE CORPORATION (1)
5,642,440 System using ergodic ensemble for image restoration 1 1994
 
JUSTSYSTEMS CORPORATION (1)
6,618,697 Method for rule-based correction of spelling and grammar errors 60 1999
 
KOREA UNIVERSITY RESEARCH AND BUSINESS FOUNDATION (1)
8,351,702 Handwriting compound system 0 2009
 
LOCKHEED MARTIN CORPORATION (1)
7,167,587 Sequential classifier for use in pattern recognition system 1 2002
 
MATSUSHITA ELECTRIC CORPORATION OF AMERICA (1)
5,768,423 Trie structure based method and apparatus for indexing and searching handwritten databases with dynamic search sequencing 67 1995
 
MITSUBISHI ELECTRIC RESEARCH LABORATORIES, INC. (1)
5,659,771 System for spelling correction in which the context of a target word in a sentence is utilized to determine which of several possible words was intended 40 1995
 
NEC CORPORATION (1)
5,774,576 Pattern recognition by unsupervised metric learning 12 1995
 
NORTHROP GRUMMAN SYSTEMS CORPORATION (1)
6,282,324 Text image deblurring by high-probability word selection 0 1995
 
PANASONIC CORPORATION OF NORTH AMERICA (1)
5,734,882 Pictographic bitmap naming of files in pen-based computer systems 19 1995
 
PERKINELMER LAS, INC. (1)
6,631,211 Interactive system for analyzing scatter plots 4 1999
 
RAMP HOLDINGS, INC. (F/K/A EVERYZING, INC.) (1)
5,621,859 Single tree method for grammar directed, very large vocabulary speech recognizer 91 1994
 
RICOH COMPANY, LTD. (1)
6,272,242 Character recognition method and apparatus which groups similar character patterns 31 1995
 
SANDIA CORPORATION (1)
6,304,675 Visual cluster analysis and pattern recognition methods 27 1993
 
SCAN-OPTICS, LLC (1)
5,850,480 OCR error correction methods and apparatus utilizing contextual comparison 69 1996
 
SHARP KABUSHIKI KAISHA (1)
5,187,751 Clustering system for optical character reader 11 1991
 
SIEMENS AKTIENGESELLSCHAFT (1)
5,949,902 Pattern recognition method which utilizes a calibration rule 2 1995
 
THE UNITED STATES OF AMERICA AS REPRESENTED BY THE SECRETARY OF THE NAVY (1)
5,825,978 Method and apparatus for speech recognition using optimized partial mixture tying of HMM state functions 34 1994
 
THOMSON REUTERS (SCIENTIFIC) INC. (1)
7,752,218 Method and apparatus for providing comprehensive search results in response to user queries entered over a computer network 0 2006
 
VADEM (1)
6,044,171 Method and apparatus for pattern recognition and representation using fourier descriptors and iterative transformation-reparametrization 21 1995
 
VISUAL PERCEPTION RESEARCH LABORATORIES (1)
6,219,449 Character recognition system 3 1996
 
OTHER [CHECK PATENT PROFILE FOR ASSIGNMENT INFORMATION] (2)
5,392,367 Automatic planar point pattern matching device and the matching method thereof 13 1993
6,434,522 Combined quantized and continuous feature vector HMM approach to speech recognition 1 1997