US Patent No: 5,621,859

Number of patents in Portfolio can not be more than 2000

Single tree method for grammar directed, very large vocabulary speech recognizer

3 Status Updates

Stats

ATTORNEY / AGENT: (SPONSORED)

Importance

Loading Importance Indicators... loading....

Abstract

See full text

The invention provides a method of large vocabulary speech recognition that employs a single tree-structured phonetic hidden Markov model (HMM) at each frame of a time-synchronous process. A grammar probability is utilized upon recognition of each phoneme of a word, before recognition of the entire word is complete. Thus, grammar probabilities are exploited as early as possible during recognition of a word. At each frame of the recognition process, a grammar probability is determined for the transition from the most likely preceding grammar state to a set of words that share at least one common phoneme. The grammar probability is combined with accumulating phonetic evidence to provide a measure of the likelihood that a state in the HMM will lead to the word most likely to have been spoken. In a preferred embodiment, phonetic context information is exploited, even before the complete context of a phoneme is known. Instead of an exact triphone model, wherein the phonemes previous and subsequent to a phoneme are considered, a composite triphone model is used that exploits partial phonetic context information to provide a phonetic model that is more accurate than aphonetic model that ignores context. In another preferred embodiment, the single phonetic tree method is used as the forward pass of a forward/backward recognition process, wherein the backward pass employs a recognition process other than the single phonetic tree method.

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

First Claim

See full text

all claims..

Related Publications

Loading Related Publications... loading....

Patent Owner(s)

Patent OwnerAddressTotal Patents
GOOGLE INC.MOUNTAIN VIEW, CA11127
GTE SERVICES CORPORATIONCAMBRIDGE, MA24
RAMP HOLDINGS, INC. (F/K/A EVERYZING, INC.)CAMBRIDGE, MA20

International Classification(s)

  • [Classification Symbol]
  • [Patents Count]

Inventor(s)

Inventor Name Address # of filed Patents Total Citations
Nguyen, Long Stoneham, MA 64 710
Schwartz, Richard M Sudbury, MA 19 911

Cited Art Landscape

Patent Info (Count) # Cites Year
 
INTERNATIONAL BUSINESS MACHINES CORPORATION (2)
4,741,036 Determination of phone weights for markov models in a speech recognition system 41 1985
4,748,670 Apparatus and method for determining a likely word sequence from labels generated by an acoustic processor 39 1985
 
KABUSHIKI KAISHA TOSHIBA (1)
5,457,768 Speech recognition apparatus using syntactic and semantic analysis 78 1992
 
MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. (1)
5,349,645 Word hypothesizer for continuous speech decoding using stressed-vowel centered bidirectional tree searches 75 1991
 
RAMP HOLDINGS, INC. (F/K/A EVERYZING, INC.) (1)
5,241,619 Word dependent N-best search method 78 1991
 
TEXAS INSTRUMENTS INCORPORATED (1)
4,984,178 Chart parser for stochastic unification grammar 68 1989
 
XEROX CORPORATION (1)
5,075,896 Character and phoneme recognition based on probability clustering 85 1989

Patent Citation Ranking

Forward Cite Landscape

Patent Info (Count) # Cites Year
 
APPLE INC. (43)
6,154,722 Method and apparatus for a speech recognition system language model that integrates a finite state grammar probability and an N-gram probability 49 1997
6,064,960 Method and apparatus for improved duration modeling of phonemes 52 1997
6,374,217 Fast update implementation for efficient latent semantic language modeling 23 1999
6,366,884 Method and apparatus for improved duration modeling of phonemes 39 1999
6,477,488 Method for dynamic context scope selection in hybrid n-gram+LSA language modeling 40 2000
6,697,779 Combined dual spectral and temporal alignment method for user authentication by voice 10 2000
6,553,344 Method and apparatus for improved duration modeling of phonemes 38 2002
6,778,952 Method for dynamic context scope selection in hybrid N-gram+LSA language modeling 38 2002
6,785,652 Method and apparatus for improved duration modeling of phonemes 3 2002
7,191,118 Method for dynamic context scope selection in hybrid N-gram+LSA language modeling 2 2004
7,403,941 System, method and technique for searching structured databases 8 2005
8,677,377 Method and apparatus for building an intelligent automated assistant 0 2006
7,720,673 Method for dynamic context scope selection in hybrid N-GRAM+LSA language modeling 0 2007
8,645,137 Fast, language-independent method for user authentication by voice 0 2007
8,620,662 Context-aware unit selection 1 2007
8,768,702 Multi-tiered voice feedback in an electronic device 0 2008
8,712,776 Systems and methods for selective text to speech synthesis 0 2008
8,583,418 Systems and methods of detecting language and natural language strings for text to speech synthesis 1 2008
8,676,904 Electronic devices with voice command and contextual data processing capabilities 0 2008
8,862,252 Audio user interface for displayless electronic device 0 2009
8,614,431 Automated response to and sensing of user activity in portable devices 1 2009
8,682,649 Sentiment prediction from textual data 1 2009
8,600,743 Noise profile determination for voice-related feature 0 2010
8,682,667 User profiling for selecting user specific voice input processing information 0 2010
8,713,021 Unsupervised document clustering using latent semantic density analysis 0 2010
8,719,006 Combined statistical and rule-based part-of-speech tagging for text-to-speech synthesis 0 2010
8,719,014 Electronic device with text error correction based on voice recognition data 0 2010
8,781,836 Hearing assistance system for providing consistent human speech 0 2011
8,812,294 Translating phrases from one language into another using an order-based set of declarative rules 0 2011
8,706,472 Method for disambiguating multiple readings in language conversion 0 2011
8,762,156 Speech recognition repair using contextual information 0 2011
8,688,446 Providing text input using speech data and non-speech data 0 2011
8,775,442 Semantic search using a single-source semantic model 0 2012
8,762,469 Electronic devices with voice command and contextual data processing capabilities 0 2012
8,713,119 Electronic devices with voice command and contextual data processing capabilities 0 2012
8,670,985 Devices and methods for identifying a prompt corresponding to a voice input in a sequence of prompts 0 2012
8,799,000 Disambiguation based on active input elicitation by intelligent automated assistant 0 2012
8,706,503 Intent deduction based on previous user interactions with voice assistant 0 2012
8,670,979 Active input elicitation by intelligent automated assistant 0 2012
8,660,849 Prioritizing selection criteria by automated assistant 1 2012
8,718,047 Text to speech conversion of text messages from mobile communication devices 0 2012
8,751,238 Systems and methods for determining the language to use for speech generated by a text to speech engine 0 2013
8,731,942 Maintaining context information between user interactions with a voice assistant 0 2013
 
MICROSOFT CORPORATION (20)
5,913,193 Method and system of runtime acoustic unit selection for speech synthesis 135 1996
5,937,384 Method and system for speech recognition using continuous density hidden Markov models 68 1996
6,904,402 System and iterative method for lexicon, segmentation and language model joint optimization 64 2000
7,451,075 Compressed speech lexicon and method and apparatus for creating and accessing the speech lexicon 1 2000
7,139,709 Middleware layer between speech related applications and engines 3 2000
6,957,184 Context free grammar engine for speech recognition system 13 2000
6,961,694 Method and apparatus for reducing latency in speech-based applications 20 2001
6,931,376 Speech-related event notification system 6 2001
7,529,671 Block synchronous decoding 34 2003
7,089,189 Speech-related event notification system 7 2004
7,162,425 Speech-related event notification system 1 2004
7,177,807 Middleware layer between speech related applications and engines 1 2004
7,177,813 Middleware layer between speech related applications and engines 0 2004
7,155,392 Context free grammar engine for speech recognition system 2 2005
7,206,742 Context free grammar engine for speech recognition system 3 2005
8,209,175 Uncertainty interval content sensing within communications 4 2006
7,379,874 Middleware layer between speech related applications and engines 50 2006
8,135,590 Position-dependent phonetic models for reliable pronunciation identification 1 2007
8,060,360 Word-dependent transition models in HMM based word alignment for statistical machine translation 15 2007
8,355,917 Position-dependent phonetic models for reliable pronunciation identification 3 2012
 
SONY ELECTRONICS INC. (4)
6,006,186 Method and apparatus for a parameter sharing speech recognition system 20 1997
6,173,258 Method for reducing noise distortions in a speech recognition system 26 1998
6,768,979 Apparatus and method for noise attenuation in a speech recognition system 30 1999
7,139,708 System and method for speech recognition using an enhanced phone set 6 1999
 
U.S. PHILIPS CORPORATION (4)
5,873,061 Method for constructing a model of a new word for addition to a word model database of a speech recognition system 28 1996
5,995,930 Method and apparatus for recognizing spoken words in a speech signal by organizing the vocabulary in the form of a tree 15 1996
6,081,779 Language model adaptation for automatic speech recognition 21 1998
6,182,026 Method and device for translating a source text into a target using modeling and dynamic programming 32 1998
 
AT&T CORP. (3)
6,233,544 Method and apparatus for language translation 85 1996
6,574,597 Fully expanded context-dependent networks for speech recognition 94 2000
7,440,893 Automated dialog method with first and second thresholds for adapted dialog strategy 7 2005
 
BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY (3)
5,819,222 Task-constrained connected speech recognition of propagation of tokens only if valid propagation path is present 5 1995
5,848,388 Speech recognition with sequence parsing, rejection and pause detection options 36 1995
5,905,971 Automatic speech recognition 8 1996
 
CANON KABUSHIKI KAISHA (3)
5,812,975 State transition model design method and voice recognition method and apparatus using same 50 1996
6,226,610 DP Pattern matching which determines current path propagation using the amount of path overlap to the subsequent time point 27 1999
7,565,290 Speech recognition method and apparatus 0 2005
 
GREAT NORTHERN RESEARCH, LLC (3)
8,165,886 Speech interface system and method for control and interaction with applications on a computing system 55 2008
8,219,407 Method for processing the output of a speech recognizer 43 2008
8,793,137 Method for processing the output of a speech recognizer 0 2012
 
MOTOROLA MOBILITY LLC (3)
6,182,038 Context dependent phoneme networks for encoding speech information 54 1997
8,316,302 Method and apparatus for annotating video content with metadata generated using speech recognition technology 1 2007
8,793,583 Method and apparatus for annotating video content with metadata generated using speech recognition technology 0 2012
 
NUANCE COMMUNICATIONS, INC. (3)
5,963,905 Method and apparatus for improving acoustic fast match speed using a cache for phone probabilities 3 1997
7,143,035 Methods and apparatus for generating dialog state conditioned language models 5 2002
7,827,031 Method for accelerating the execution of speech recognition neural networks and the related speech recognition device 0 2003
 
ORACLE INTERNATIONAL CORPORATION (3)
8,396,859 Subject matter context search engine 0 2004
8,190,985 Frame-slot architecture for data conversion 0 2009
8,832,075 Subject matter context search engine 0 2013
 
SONY ONLINE ENTERTAINMENT LLC (3)
8,050,924 System for generating and selecting names 0 2005
7,912,716 Generating words and names using N-grams of phonemes 1 2005
8,359,200 Generating profiles of words 0 2011
 
AT&T INTELLECTUAL PROPERTY I, L.P. (2)
8,548,807 System and method for adapting automatic speech recognition pronunciation by acoustic model restructuring 1 2009
8,812,315 System and method for adapting automatic speech recognition pronunciation by acoustic model restructuring 0 2013
 
CASTELL SOFTWARE LIMITED LIABILITY COMPANY (2)
7,085,717 Scoring and re-scoring dynamic time warping of speech 4 2002
6,983,246 Dynamic time warping using frequency distributed distance measures 6 2002
 
INTEL CORPORATION (2)
7,346,495 Method and system for building a domain specific statistical language model from rule based grammar specifications 6 2000
6,980,954 Search method based on single triphone tree for large vocabulary continuous speech recognizer 7 2000
 
INTERNATIONAL BUSINESS MACHINES CORPORATION (2)
7,398,274 Mention-synchronous entity tracking system and method for chaining mentions 0 2004
8,620,961 Mention-synchronous entity tracking: system and method for chaining mentions 0 2008
 
VERIZON PATENT AND LICENSING INC. (2)
7,092,888 Unsupervised training in natural language call routing 72 2002
7,478,043 Estimation of speech spectral parameters in the presence of noise 1 2003
 
Accumente, LLC (1)
8,886,535 Utilizing multiple processing units for rapid training of hidden markov models 0 2014
 
ADACEL SYSTEMS, INC. (1)
8,515,734 Integrated language model, related systems and methods 0 2010
 
ALCATEL-LUCENT USA INC. (1)
5,832,430 Devices and methods for speech recognition of vocabulary words with simultaneous detection and verification 63 1995
 
ASAHI KASEI KABUSHIKI KAISHA (1)
7,272,561 Speech recognition device and speech recognition method 4 2001
 
AT&T INTELLECTUAL PROPERTY II, L.P. (1)
7,487,088 Method and system for predicting understanding errors in a task classification system 36 2006
 
CALABRIO, INC. (1)
8,543,393 Systems and methods of improving automated speech recognition accuracy using statistical analysis of search terms 0 2008
 
CISCO TECHNOLOGY, INC. (1)
6,230,128 Path link passing speech recognition with vocabulary node being capable of simultaneously processing plural path links 7 1995
 
DENSO CORPORATION (1)
7,818,171 Speech recognition apparatus and speech recognition program 1 2007
 
FONDAZIONE BRUNO KESSLER (1)
5,765,133 System for building a language model network for speech recognition 27 1996
 
GOOGLE INC. (1)
8,768,712 Initiating actions based on partial hotwords 0 2013
 
HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P. (1)
5,819,220 Web triggered word set boosting for speech interfaces to the world wide web 192 1996
 
LONGSAND LIMITED (1)
5,983,180 Recognition of sequential data using finite state sequence models organized in a tree structure 38 1998
 
LUCENT TECHNOLOGIES INC. (1)
5,870,706 Method and apparatus for an improved language recognition system 80 1996
 
MASSACHUSETTS INSTITUTE OF TECHNOLOGY (1)
6,317,716 Automatic cueing of speech 26 1998
 
MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. (1)
7,013,273 Speech recognition based captioning system 15 2001
 
MITSUBISHI ELECTRIC RESEARCH LABORATORIES, INC. (1)
7,171,358 Compression of language model structures and word identifiers for automated speech recognition systems 1 2003
 
NIPPON TELEGRAPH AND TELEPHONE CORPORATION (1)
5,835,890 Method for speaker adaptation of speech models recognition scheme using the method and recording medium having the speech recognition method recorded thereon 44 1997
 
NOKIA CORPORATION (1)
7,319,960 Speech recognition method and system 5 2001
 
POPKIN FAMILY ASSETS, L.L.C. (1)
6,397,179 Search optimization system and method for continuous speech recognition 40 1998
 
RAMP HOLDINGS, INC. (F/K/A EVERYZING, INC.) (1)
6,052,682 Method of and apparatus for recognizing and labeling instances of name classes in textual environments 27 1997
 
RENESAS ELECTRONICS CORPORATION (1)
6,112,173 Pattern recognition device using tree structure data 4 1998
 
ROCKSTAR CONSORTIUM US LP (1)
6,092,045 Method and apparatus for speech recognition 38 1998
 
SAMSUNG ELECTRONICS CO., LTD. (1)
8,849,668 Speech recognition apparatus and method 0 2011
 
SCANSOFT, INC. (1)
6,275,802 Search algorithm for large vocabulary speech recognition 7 1999
 
SENSORY, INCORPORATED (1)
8,700,399 Systems and methods for hands-free voice control and voice search 0 2010
 
SONY CORPORATION (1)
5,787,395 Word and pattern recognition through overlapping hierarchical tree defined by relational features 14 1996
 
US PHILIPS ELECTRONICS (1)
7,006,971 Recognition of a speech utterance available in spelled form 7 2000
 
VICTOR COMPANY OF JAPAN, LTD. (1)
5,799,277 Acoustic model generating method for speech recognition 13 1995
 
VOLT DELTA RESOURCES LLC (1)
5,987,414 Method and apparatus for selecting a vocabulary sub-set from a speech recognition dictionary for use in real time automated directory assistance 30 1996
 
Other [Check patent profile for assignment information] (2)
8,898,568 Audio user interface 0 2008
8,892,446 Service orchestration for intelligent automated assistant 0 2012