US20040148165A1 - Pattern processing system specific to a user group - Google Patents

Pattern processing system specific to a user group Download PDF

Info

Publication number
US20040148165A1
US20040148165A1 US10/479,554 US47955403A US2004148165A1 US 20040148165 A1 US20040148165 A1 US 20040148165A1 US 47955403 A US47955403 A US 47955403A US 2004148165 A1 US2004148165 A1 US 2004148165A1
Authority
US
United States
Prior art keywords
user
user group
specific
pattern processing
group
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US10/479,554
Other languages
English (en)
Inventor
Peter Beyerlein
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nuance Communications Inc
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Assigned to KONINKLIJKE PHILIPS ELECTRONICS N.V. reassignment KONINKLIJKE PHILIPS ELECTRONICS N.V. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: BEYERLEIN, PETER
Publication of US20040148165A1 publication Critical patent/US20040148165A1/en
Assigned to NUANCE COMMUNICATIONS, INC. reassignment NUANCE COMMUNICATIONS, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: KONINKLIJKE PHILIPS ELECTRONICS N.V.
Priority to US13/589,394 priority Critical patent/US9009043B2/en
Priority to US14/637,049 priority patent/US9424838B2/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
    • G10L15/065—Adaptation
    • G10L15/07—Adaptation to the speaker

Definitions

  • the invention relates to a pattern processing system and in particular to a speech processing system.
  • Pattern processing systems and in particular those with speech recognition are used in many locations and for many applications.
  • Examples are the automatic information and transaction systems which are available by telephone, for example the automatic timetable information of the Dutch public transport organizations (OVR) or the telebanking systems of many banks, as well as the information kiosks of the Philips company put up in the city of Vienna, where a user can obtain, for example, information on the sites and the hotels of Vienna by means of keyboard and spoken inputs.
  • OVR automatic timetable information of the Dutch public transport organizations
  • telebanking systems of many banks as well as the information kiosks of the Philips company put up in the city of Vienna, where a user can obtain, for example, information on the sites and the hotels of Vienna by means of keyboard and spoken inputs.
  • pattern processing systems are to be used by many users, so-termed user-independent pattern processing data sets are mostly used for the pattern processing, i.e. no distinction is made between the users in the processing of patterns from different users; for example, the same acoustic reference models are used in speech recognition for all users. It is known to those skilled in the art, however, that the quality of pattern processing is improved through the use of user-specific pattern processing data sets. For example, the accuracy of speech recognition systems is enhanced if a standardization of vowel lengths specially attuned to a given speaker is carried out for the spoken utterances of this speaker.
  • Such speaker-dependent speech recognition systems are widely used nowadays in applications with small user numbers.
  • Examples are personal dictation systems, for example FreeSpeech of the Philips company, or the professional dictation systems for closed user groups, for example SpeechMagic of the Philips company in the field of X-ray medicine.
  • a transfer of these techniques to pattern processing systems with many users, however, is hampered by many difficulties.
  • the problem of the high storage requirement is avoided through the subdivision of the users into user groups.
  • the pattern processing data sets specific to the user groups may be accommodated in central data memories and may be made available to the pattern processing devices through a network. Further possibilities for saving memory space thus arise from the avoidance of multiple data storage.
  • the multiply shared use of the pattern processing data sets specific to user groups in a plurality of systems avoids the problems involved in a multiple laying-down of the user group for the user.
  • a user wants to use the pattern processing system specific to the user group, he must inform the system only of his user group, for example by means of a number or a symbolic name.
  • the user group information may also be accommodated on a chip card, but for using, for example, a telephonic information system it also suffices to inform the system verbally of the user group or, for example, to enter the number through the keyboard of a DTMF-capable telephone in the case of a numerical code.
  • the respective pattern processing system specific to the user group may thus be used also without additional equipment such as, for example, an information card and also by means of existing user terminals such as, for example, a telephone.
  • the user group may be defined for a user in a training phase, as claimed in claim 2 , in which the user, for example, has to pronounce a given text which is recorded by a training system and is used for determining the user group.
  • This training phase may take place independently of a use of a pattern processing system specific to a user group. Alternatively, however, it may be offered during the use of the system to a “new” user, i.e. a user with whom no user group has yet been associated. It is furthermore conceivable to use the pattern inputs of the user entered during use of the system, which were perhaps initially treated with a user-independent pattern processing, for the first or alternatively the renewed definition of the user group. The latter may occur when the pattern characteristics of the user or the user groups of the system have changed.
  • Claim 3 defines how the definition of the user group for the user can be influenced by the user. It is conceivable here, for example, that a system offers user groups of different qualities. Thus a system may offer, for example, user groups of high quality in that it offers very finely distinguished acoustic reference models for these groups, in which exclusively users of very similar speaking and behavior modes will be present. As a result, recognition accuracies may be offered to such a user group, for example in speech recognition, similar to those obtained in user-dependent systems. The higher expenditure necessary for this purpose in the system could be recovered from the user by means of a corresponding tariff structure.
  • the dependent claims 4 and 5 relate to two advantageous possibilities for user input.
  • user inputs may be made into a public user terminal such as, for example, an information kiosk as mentioned above or an automatic bank counter.
  • a user may use a telephone or PC or laptop, in which case his inputs are transmitted via a network, for example the telephone network or the Internet.
  • the dependent claim 6 specifies a few possible components of a pattern processing data set specific to a user group:
  • acoustic reference models may be available, for example, in the form of so-termed Hidden Markov Models for the speech sounds of a language.
  • Vocabularies specific to a user group contain, for example, the words typically used by a user group for an application.
  • Language models may comprise all interrelationships for the formation of a word sequence, i.e., for example, also grammatical rules or semantic preferences of the user group, while dialogue models identify the characteristic patterns of interaction between the system and the users from a user group.
  • the invention also relates to other types of pattern processing specific to user groups such as, for example, speech encoding specific to a user group, for example through the use of code books specific to a user group.
  • Handwriting recognition and facial expression processing specific to a user group for example in systems for on-line chatting with animated characters, so-called avatars, also fall within the scope of the invention.
  • Claim 7 claims the use of the system for providing information such as, for example, timetable or touristic information. It is furthermore claimed to use the system for the issuing of orders such as, for example, for effecting purchases from an automatic vending machine or carrying out bank transactions via the Internet.
  • FIGS. 1 and 2 show embodiments of the pattern processing system according to the invention which is specific to a user group
  • FIG. 3 diagrammatically shows the contents of a data memory for the pattern processing data sets specific to the user group
  • FIG. 4 shows the sequence of use of a pattern processing system specific to the user group according to the invention in the form of a flowchart.
  • FIG. 1 shows an embodiment of the pattern processing system specific to a user group according to the invention which provides a public user terminal 10 for user inputs.
  • Typical applications of such an embodiment of the invention are the carrying out of bank transactions at an automatic bank counter 10 , calling up of information at the information kiosks 10 mentioned above, or purchasing tickets from an automatic ticket vending machine 10 .
  • the public user terminal 10 has a graphic display 11 , an input keyboard 12 , a microphone 13 for putting in spoken messages of a user, and an insertion slot 14 for a chip card 70 characteristic of the user, for example an EC or customer card. Furthermore, it comprises a local speech recognition device 42 which has a local data memory 32 for storing pattern processing data sets specific to a user group for a speech processing specific to the user group.
  • a chip 71 for storing data typical of the application for example the account number of the bank account in the case of a bank card
  • Such a unique identifier may consist, for example, of a number, or alternatively a symbolic name is conceivable, for example the name of a widely known person who also belongs to this user group such that the speech processing characteristic of this person is typical of the user group.
  • a symbolic name but also a number can be readily remembered by a user so that such a unique identifier can be transmitted to the pattern processing system also without the help of a chip card, for example through the microphone 13 or the keyboard 12 .
  • the chip 72 would be redundant on the chip card 70 .
  • all information could be accommodated on a single chip 71 on the chip card 70 , including any user group characterization.
  • the local speech recognition device 42 is capable of operating fully independently locally.
  • Such a “stand alone” automatic machine is particularly suitable, for example, for the sale of cigarettes or some other articles directly available from vending machines.
  • the public user terminal 10 may alternatively be connected via a network 20 to further data memories 30 . . . 31 for the storage of pattern processing data sets specific to user groups for the purpose of a speech recognition specific to the respective user group.
  • the network 20 may then be, for example, a private MAN (Metropolitan Area Network), for example the network of a bank.
  • the network 20 may alternatively be realized in the form of a public network, in particular the Internet.
  • a possible hybrid form is, for example, a VPN (Virtual Private Network) realized on the basis of the Internet.
  • FIG. 2 shows a further embodiment of the pattern processing system according to the invention which is specific to a user group.
  • the network 20 and the data memories 30 . . . 31 connected thereto are shown.
  • the speech recognition devices 40 . . . 41 are also connected to the network 20 .
  • the user inputs are made in a public user terminal 10 here, which terminal, unlike FIG. 1, has no local speech recognition device 42 with a local data memory 32 , or through a telephone 60 or a PC, a laptop, or the like 50 , which are all connected to the network 20 for this purpose or can be connected thereto.
  • These and other input possibilities such as, for example, the public user terminal 10 shown in FIG. 1 with a local speech recognition device may be realized all or only in part in a pattern processing system specific to a user group.
  • FIGS. 1 and 2 accordingly differ especially in the arrangement of the speech recognition device 42 or 40 . . . 41 in which the recognition of the spoken utterances of a user takes place.
  • the speech recognition device 42 accommodated locally in the public user terminal 10 of FIG. 1 is particularly suitable for the case in which only simple commands are to be recognized and the public user terminal 10 is mainly used by the same customers all the time.
  • a comparatively simple and inexpensive speech recognition device 42 will suffice in this case, and the pattern processing data sets specific to the user group of the main users may be stored in the local data memory 32 of the speech recognition device 42 for a speech recognition which is specific to the user group.
  • Further pattern processing data sets specific to the user group are loaded from data memories 30 . . . 31 connected via the network 20 . This leads to a small overall load of the network 20 .
  • the speech recognition of the spoken utterances of a user takes place in the speech recognition devices 40 . . . 41 connected via the network.
  • This is useful in the case of more complicated spoken utterances which require a high recognition performance and/or in the case of continually changing users.
  • the joining together of the speech recognition tasks and the data storage yields advantages in the machine occupancy, the memory space required, and the necessary data traffic through the network 20 .
  • it may be useful, for example, to connect the speech recognition devices 40 . . . 41 with one another and with the data memories 30 . . . 31 by means of a broadband sub-network within the network 20 .
  • a further embodiment of the invention arises when the pattern processing data sets specific to user groups are not held in data memories belonging to a system and designed for pattern processing specific to the user groups, but are made available, for example, by a third-party provider or alternatively by a user himself (for his own user group).
  • third parties may specialize in the creation, management, and/or updating of the pattern processing data sets specific to the user groups in order to make them available to the operators of the pattern processing systems specific to the user groups, for example against payment. Third parties may also take care of the definition of the user group membership for the users.
  • a user himself would download the pattern processing data sets specific to his own user group, for example from one of the data memories 32 , 30 . . . 31 of a pattern processing system specific to the user group.
  • a different pattern processing system specific to user groups which system itself does not have the pattern processing data sets specific to the respective user group of this user, he may make the respective data available to the system on the laptop 50 .
  • the message containing the address of the PC or laptop 50 would perform the task of providing the unique identifier of the user group.
  • User terminals used in the embodiments described above for obtaining access to the system were public user terminals 10 of average complexity, telephones 60 , and PCs or laptops 50 , but alternative solutions are equally possible. Examples are mobile telephones and information kiosks with complicated multimedia interaction possibilities such as touch screens, cameras, loudspeakers, etc.
  • FIG. 3 is a diagram showing the contents of a data memory 30 for the pattern processing data sets 80 . . . 81 specific to the user groups.
  • the data memory 30 which here represents the local data memory 32 as well as the further data memories 30 . . . 31 connected to the network 20 , is a known computer data memory, for example a hard disk.
  • the pattern processing data sets 80 . . . 81 specific to user groups may be available in the form of individual data files, for example in binary code suitable for the pattern processing system specific to user groups.
  • An alternative possibility is an organization in the form of a database or the like.
  • FIG. 4 shows a possible sequence of the use of a pattern processing system specific to user groups according to the invention in the form of a flowchart. Only those processes are discussed which are relevant to the pattern processing specific to user groups, while actions specific to the application such as, for example, the communication of a bank account number and a PIN code for a banking application are not represented.
  • the pattern processing system specific to user groups requests a user in process block 102 to identify his user group, i.e. to enter into the system a unique identifier of the user group defined for the respective user for the purpose of a pattern processing specific to the user group.
  • the decision block 103 the further process branches off in dependence on whether the user knows his user group or not.
  • the user knows his own user group, he communicates it to the system in block 104 in that, for example, he inserts the chip card 70 into the insertion slot 14 of a public user terminal 10 , in the scenarios of FIGS. 1 and 2, he uses the keyboard 12 or the microphone 13 of the public user terminal 10 , or he makes the user group known to the system through a telephone 60 or a laptop 50 .
  • the system searches in block 105 the pattern processing data set specific to the user group of the user in a data memory 32 , 30 . . . 31 and makes it available in a pattern processing device 42 , 40 . . . 41 .
  • the system requests him in block 106 whether he wants the system to define a user group for him now. If he wants to do so, the system collects training pattern inputs of the user in block 107 and processes these so as to define a user group for the user. The user group thus determinesd is communicated to the user in block 108 , and the control switches to block 105 described above, in which the pattern processing data set specific to the user group of the current user is looked up in a data memory 32 , 30 . . . 31 and is made available to a pattern processing device 42 , 40 . . . 41 .
  • the control branches from block 106 to block 109 .
  • There a user-independent pattern processing data set is looked up in a data memory 32 , 30 . . . 31 and is made available to a pattern processing device 42 , 40 . . . 41 such that the subsequent pattern processing steps are carried out independently of the special characteristics of the user.
  • the user-group-specific or user-independent pattern processing data set made available to the pattern processing device 42 , 40 . . . 41 in one of the blocks 105 and 109 may be dependent on further conditions.
  • different ambient conditions may hold for different applications, for example different background noises in the case of speech recognition, or different terminals for user inputs such as the microphone type in the case of speech input or the camera type in the case of gesture recognition, and a suitably adapted pattern processing data set may be used for these.
  • the pattern input of the user is processed in block 110 , i.e. the user is requested to enter a pattern and the pattern entered is recorded and processed.
  • Such pattern inputs may be spoken utterances put in through a microphone 13 or a telephone 60 .
  • Other possible inputs are handwritten texts and/or pointer actions for selecting menu items offered on the display 11 .
  • the display 11 of the public user terminal 10 may be constructed for this purpose, for example, as a touch screen, and/or the public user terminal 10 could be fitted with a camera.
  • the pattern inputs of the user effected in block 110 may be put into intermediate storage and may be used, for example, for testing the user group definition for the user.
  • the system may load a better suited pattern processing data set specific to the user group of the user, in consultation with the user, into a pattern processing device 42 , 40 . . . 41 so as to carry out the further pattern processing steps therewith.
  • Such a procedure may also be carried out, for example, if the patterns had been processed up to that moment on the basis of a user-independent pattern processing data set.
  • the actions corresponding to the pattern input of the user are carried out, for example account data are shown on the display 11 of the public user terminal 10 in the case of a bank transaction. It is also possible for return questions to be put to the user. The user may also be requested for a further input such as, for example, a lacking bank code number.
  • the termination of the interaction with the user in block 112 may follow, for example, the recognition of a positive reply of the user to a relevant previous system question in block 110 .
  • a termination button on the input keyboard 12 of the public user terminal 10 may be provided, which may be operated at any moment in the man-machine communication. Further modifications obvious to those skilled in the art are conceivable.
  • Blocks 107 and 113 provided the possibility of defining a user group for the user during such a man-machine communication, and blocks 110 and 113 rendered it possible to modify such a user group definition. Defining or modifying a user group, however, need not take place within the framework of a utilization of the system, for example for carrying out bank transactions, but it may alternatively be done separately.
  • FIG. 4 clarified essential aspects of a method according to the invention for a pattern processing specific to user groups, it will be clear to those skilled in the art that such a method should contain further mechanisms in practice, for example for the treatment of error conditions.
  • the user group of a user laid down by a system is not known to another system.
  • This other system may then act for the purpose of error treatment, for example, exactly as in the case described starting from block 106 , where the user does not know his own user group at that particular moment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Collating Specific Patterns (AREA)
  • Silicon Polymers (AREA)
  • Developing Agents For Electrophotography (AREA)
  • Macromonomer-Based Addition Polymer (AREA)
US10/479,554 2001-06-06 2002-06-05 Pattern processing system specific to a user group Abandoned US20040148165A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/589,394 US9009043B2 (en) 2001-06-06 2012-08-20 Pattern processing system specific to a user group
US14/637,049 US9424838B2 (en) 2001-06-06 2015-03-03 Pattern processing system specific to a user group

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
DE10127559.5 2001-06-06
DE2001127559 DE10127559A1 (de) 2001-06-06 2001-06-06 Benutzergruppenspezifisches Musterverarbeitungssystem
PCT/IB2002/002055 WO2002099785A1 (en) 2001-06-06 2002-06-05 Pattern processing system specific to a user group

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2002/002055 A-371-Of-International WO2002099785A1 (en) 2001-06-06 2002-06-05 Pattern processing system specific to a user group

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US13/589,394 Continuation US9009043B2 (en) 2001-06-06 2012-08-20 Pattern processing system specific to a user group

Publications (1)

Publication Number Publication Date
US20040148165A1 true US20040148165A1 (en) 2004-07-29

Family

ID=7687445

Family Applications (3)

Application Number Title Priority Date Filing Date
US10/479,554 Abandoned US20040148165A1 (en) 2001-06-06 2002-06-05 Pattern processing system specific to a user group
US13/589,394 Expired - Fee Related US9009043B2 (en) 2001-06-06 2012-08-20 Pattern processing system specific to a user group
US14/637,049 Expired - Fee Related US9424838B2 (en) 2001-06-06 2015-03-03 Pattern processing system specific to a user group

Family Applications After (2)

Application Number Title Priority Date Filing Date
US13/589,394 Expired - Fee Related US9009043B2 (en) 2001-06-06 2012-08-20 Pattern processing system specific to a user group
US14/637,049 Expired - Fee Related US9424838B2 (en) 2001-06-06 2015-03-03 Pattern processing system specific to a user group

Country Status (6)

Country Link
US (3) US20040148165A1 (de)
EP (1) EP1402518B1 (de)
JP (1) JP4837887B2 (de)
AT (1) ATE340399T1 (de)
DE (2) DE10127559A1 (de)
WO (1) WO2002099785A1 (de)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050251287A1 (en) * 2004-05-05 2005-11-10 Provision Interactive Technologies, Inc. System and method for dispensing consumer products
WO2007132404A2 (en) 2006-05-12 2007-11-22 Koninklijke Philips Electronics N.V. Method for changing over from a first adaptive data processing version to a second adaptive data processing version
US20080167871A1 (en) * 2007-01-04 2008-07-10 Samsung Electronics Co., Ltd. Method and apparatus for speech recognition using device usage pattern of user
CN103544337A (zh) * 2012-05-29 2014-01-29 通用汽车环球科技运作有限责任公司 汽车乘员的对话模型
CN104412322A (zh) * 2012-06-29 2015-03-11 埃尔瓦有限公司 用于管理适应数据的方法和系统
US9620128B2 (en) 2012-05-31 2017-04-11 Elwha Llc Speech recognition adaptation systems based on adaptation data
US9899040B2 (en) 2012-05-31 2018-02-20 Elwha, Llc Methods and systems for managing adaptation data
US9899026B2 (en) 2012-05-31 2018-02-20 Elwha Llc Speech recognition adaptation systems based on adaptation data
US10431235B2 (en) 2012-05-31 2019-10-01 Elwha Llc Methods and systems for speech adaptation data
US20220005481A1 (en) * 2018-11-28 2022-01-06 Samsung Electronics Co., Ltd. Voice recognition device and method

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030220788A1 (en) * 2001-12-17 2003-11-27 Xl8 Systems, Inc. System and method for speech recognition and transcription
KR101619262B1 (ko) * 2014-11-14 2016-05-18 현대자동차 주식회사 음성인식 장치 및 방법
KR20170034227A (ko) * 2015-09-18 2017-03-28 삼성전자주식회사 음성 인식 장치 및 방법과, 음성 인식을 위한 변환 파라미터 학습 장치 및 방법
US10268683B2 (en) * 2016-05-17 2019-04-23 Google Llc Generating output for presentation in response to user interface input, where the input and/or the output include chatspeak
TWI682386B (zh) * 2018-05-09 2020-01-11 廣達電腦股份有限公司 整合式語音辨識系統及方法
JP7261096B2 (ja) * 2019-06-13 2023-04-19 株式会社日立製作所 計算機システム、モデル生成方法及びモデル管理プログラム

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6167377A (en) * 1997-03-28 2000-12-26 Dragon Systems, Inc. Speech recognition language models
US6253181B1 (en) * 1999-01-22 2001-06-26 Matsushita Electric Industrial Co., Ltd. Speech recognition and teaching apparatus able to rapidly adapt to difficult speech of children and foreign speakers
US20020065656A1 (en) * 2000-11-30 2002-05-30 Telesector Resources Group, Inc. Methods and apparatus for generating, updating and distributing speech recognition models
US6442519B1 (en) * 1999-11-10 2002-08-27 International Business Machines Corp. Speaker model adaptation via network of similar users
US6493669B1 (en) * 2000-05-16 2002-12-10 Delphi Technologies, Inc. Speech recognition driven system with selectable speech models
US6665639B2 (en) * 1996-12-06 2003-12-16 Sensory, Inc. Speech recognition in consumer electronic products
US6735563B1 (en) * 2000-07-13 2004-05-11 Qualcomm, Inc. Method and apparatus for constructing voice templates for a speaker-independent voice recognition system
US6873951B1 (en) * 1999-03-30 2005-03-29 Nortel Networks Limited Speech recognition system and method permitting user customization
US7103549B2 (en) * 2001-03-22 2006-09-05 Intel Corporation Method for improving speech recognition performance using speaker and channel information

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6101468A (en) * 1992-11-13 2000-08-08 Dragon Systems, Inc. Apparatuses and methods for training and operating speech recognition systems
JPH08123461A (ja) * 1994-10-20 1996-05-17 Hitachi Ltd 個人情報カードを用いた音声インタフェースシステム
US5895447A (en) * 1996-02-02 1999-04-20 International Business Machines Corporation Speech recognition using thresholded speaker class model selection or model adaptation
US6182037B1 (en) * 1997-05-06 2001-01-30 International Business Machines Corporation Speaker recognition over large population with fast and detailed matches
JP2000089780A (ja) * 1998-09-08 2000-03-31 Seiko Epson Corp 音声認識方法および音声認識装置
US7505905B1 (en) * 1999-05-13 2009-03-17 Nuance Communications, Inc. In-the-field adaptation of a large vocabulary automatic speech recognizer (ASR)
JP2000347684A (ja) * 1999-06-02 2000-12-15 Internatl Business Mach Corp <Ibm> 音声認識システム
EP1134725A1 (de) * 2000-03-15 2001-09-19 Siemens Aktiengesellschaft Anpassung automatischer Spracherkennungssysteme an spezifische Charakteristika von einzelnen Sprechergruppen zur Verbesserung der Erkennungsleistung
US20020046030A1 (en) * 2000-05-18 2002-04-18 Haritsa Jayant Ramaswamy Method and apparatus for improved call handling and service based on caller's demographic information
US7272561B2 (en) * 2000-07-13 2007-09-18 Asahi Kasei Kabushiki Kaisha Speech recognition device and speech recognition method
DE10047724A1 (de) * 2000-09-27 2002-04-11 Philips Corp Intellectual Pty Verfahren zur Ermittlung eines Eigenraumes zur Darstellung einer Mehrzahl von Trainingssprechern
DE10047718A1 (de) * 2000-09-27 2002-04-18 Philips Corp Intellectual Pty Verfahren zur Spracherkennung
US6785647B2 (en) * 2001-04-20 2004-08-31 William R. Hutchison Speech recognition system with network accessible speech processing resources

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6665639B2 (en) * 1996-12-06 2003-12-16 Sensory, Inc. Speech recognition in consumer electronic products
US6167377A (en) * 1997-03-28 2000-12-26 Dragon Systems, Inc. Speech recognition language models
US6253181B1 (en) * 1999-01-22 2001-06-26 Matsushita Electric Industrial Co., Ltd. Speech recognition and teaching apparatus able to rapidly adapt to difficult speech of children and foreign speakers
US6873951B1 (en) * 1999-03-30 2005-03-29 Nortel Networks Limited Speech recognition system and method permitting user customization
US6442519B1 (en) * 1999-11-10 2002-08-27 International Business Machines Corp. Speaker model adaptation via network of similar users
US6493669B1 (en) * 2000-05-16 2002-12-10 Delphi Technologies, Inc. Speech recognition driven system with selectable speech models
US6735563B1 (en) * 2000-07-13 2004-05-11 Qualcomm, Inc. Method and apparatus for constructing voice templates for a speaker-independent voice recognition system
US20020065656A1 (en) * 2000-11-30 2002-05-30 Telesector Resources Group, Inc. Methods and apparatus for generating, updating and distributing speech recognition models
US7103549B2 (en) * 2001-03-22 2006-09-05 Intel Corporation Method for improving speech recognition performance using speaker and channel information

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7881822B2 (en) * 2004-05-05 2011-02-01 Provision Interactive Technologies, Inc. System and method for dispensing consumer products
US20050251287A1 (en) * 2004-05-05 2005-11-10 Provision Interactive Technologies, Inc. System and method for dispensing consumer products
US9009695B2 (en) 2006-05-12 2015-04-14 Nuance Communications Austria Gmbh Method for changing over from a first adaptive data processing version to a second adaptive data processing version
CN101443732A (zh) * 2006-05-12 2009-05-27 皇家飞利浦电子股份有限公司 用于从第一自适应数据处理版本转换至第二自适应数据处理版本的方法
WO2007132404A2 (en) 2006-05-12 2007-11-22 Koninklijke Philips Electronics N.V. Method for changing over from a first adaptive data processing version to a second adaptive data processing version
US20090125899A1 (en) * 2006-05-12 2009-05-14 Koninklijke Philips Electronics N.V. Method for changing over from a first adaptive data processing version to a second adaptive data processing version
US9824686B2 (en) * 2007-01-04 2017-11-21 Samsung Electronics Co., Ltd. Method and apparatus for speech recognition using device usage pattern of user
US20080167871A1 (en) * 2007-01-04 2008-07-10 Samsung Electronics Co., Ltd. Method and apparatus for speech recognition using device usage pattern of user
US10529329B2 (en) 2007-01-04 2020-01-07 Samsung Electronics Co., Ltd. Method and apparatus for speech recognition using device usage pattern of user
CN103544337A (zh) * 2012-05-29 2014-01-29 通用汽车环球科技运作有限责任公司 汽车乘员的对话模型
US10431235B2 (en) 2012-05-31 2019-10-01 Elwha Llc Methods and systems for speech adaptation data
US9899040B2 (en) 2012-05-31 2018-02-20 Elwha, Llc Methods and systems for managing adaptation data
US9899026B2 (en) 2012-05-31 2018-02-20 Elwha Llc Speech recognition adaptation systems based on adaptation data
US10395672B2 (en) 2012-05-31 2019-08-27 Elwha Llc Methods and systems for managing adaptation data
US9620128B2 (en) 2012-05-31 2017-04-11 Elwha Llc Speech recognition adaptation systems based on adaptation data
CN104412322B (zh) * 2012-06-29 2019-01-18 埃尔瓦有限公司 用于管理适应数据的方法和系统
CN104412322A (zh) * 2012-06-29 2015-03-11 埃尔瓦有限公司 用于管理适应数据的方法和系统
US20220005481A1 (en) * 2018-11-28 2022-01-06 Samsung Electronics Co., Ltd. Voice recognition device and method
US11961522B2 (en) * 2018-11-28 2024-04-16 Samsung Electronics Co., Ltd. Voice recognition device and method

Also Published As

Publication number Publication date
US9009043B2 (en) 2015-04-14
JP2004529390A (ja) 2004-09-24
DE60214850D1 (de) 2006-11-02
US20150179164A1 (en) 2015-06-25
JP4837887B2 (ja) 2011-12-14
ATE340399T1 (de) 2006-10-15
WO2002099785A1 (en) 2002-12-12
US20120310647A1 (en) 2012-12-06
US9424838B2 (en) 2016-08-23
EP1402518B1 (de) 2006-09-20
EP1402518A1 (de) 2004-03-31
DE10127559A1 (de) 2002-12-12
DE60214850T2 (de) 2007-05-10

Similar Documents

Publication Publication Date Title
US9424838B2 (en) Pattern processing system specific to a user group
EP2109097B1 (de) Verfahren zur Personalisierung eines Dienstes
CN107481720B (zh) 一种显式声纹识别方法及装置
US10534623B2 (en) Systems and methods for providing a virtual assistant
US9571645B2 (en) Systems and methods for providing a virtual assistant
US9804820B2 (en) Systems and methods for providing a virtual assistant
US9479931B2 (en) Systems and methods for providing a virtual assistant
US20050033582A1 (en) Spoken language interface
US20060004575A1 (en) Method and apparatus for creation and user-customization of speech-enabled services
US20040120472A1 (en) Voice response system
US20050080629A1 (en) Multi-mode interactive dialogue apparatus and method
GB2409087A (en) Computer generated prompting
CN111475020A (zh) 一种信息的交互方法、交互装置、电子设备及存储介质
CN109684455A (zh) 智能客服系统的实现方法及装置、设备、可读存储介质
EP4123477A1 (de) Empfehlung von multimedia-informationen
CN108924218A (zh) 用于推送信息的方法和装置
US20020111786A1 (en) Everyday language-based computing system and method
CN112885371A (zh) 音频脱敏的方法、装置、电子设备以及可读存储介质
Kouroupetroglou et al. Speech-enabled e-Commerce for disabled and elderly persons
US20060031853A1 (en) System and method for optimizing processing speed to run multiple dialogs between multiple users and a virtual agent
Goldman et al. Voice Portals—Where Theory Meets Practice
CN120541304A (zh) 基于ai的聊天内容推荐方法及相关装置
CN121053975A (zh) 语音应答方法、语音应答系统、客户端和服务器端
CN120611042A (zh) 对话摘要生成方法、装置、电子设备、介质及程序产品
CN121053974A (zh) 语音应答方法、语音应答系统、客户端和服务器端

Legal Events

Date Code Title Description
AS Assignment

Owner name: KONINKLIJKE PHILIPS ELECTRONICS N.V., NETHERLANDS

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BEYERLEIN, PETER;REEL/FRAME:015253/0386

Effective date: 20030123

AS Assignment

Owner name: NUANCE COMMUNICATIONS, INC., MASSACHUSETTS

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:KONINKLIJKE PHILIPS ELECTRONICS N.V.;REEL/FRAME:028481/0354

Effective date: 20110720

STCB Information on status: application discontinuation

Free format text: ABANDONED -- AFTER EXAMINER'S ANSWER OR BOARD OF APPEALS DECISION