公开(公告)号：JP2010257452A

公开(公告)日：2010-11-11

申请号：JP2010093440

申请日：2010-04-14

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： MENGUSOGLU ERHAN ,  PICKERING JOHN BRIAN

IPC: G06F9/44 ,  G06F17/27

CPC classification number: G06F17/2705 ,  G06F8/10 ,  G10L15/183 ,  G10L15/193

Abstract: PROBLEM TO BE SOLVED: To provide a method of automatically extracting a meta-model language model for modeling a system for the system, a device, and software from natural language specifications of the system.
SOLUTION: A syntactic structure for representing one or more syntactic subjects of one set from the natural language specifications of the system is extracted, and first mapping between the respective meta-model elements for the syntactic subjects of the predetermined one set and system modeling meta-model language model is generated. One or more meta-model elements are generated by following the mapping for each identified syntactic subject, and the generated meta-model structure elements are provided for transformation to the model for the system.
COPYRIGHT: (C)2011,JPO&INPIT

Abstract translation: 要解决的问题：提供一种自动提取元模型语言模型的方法，用于为系统的系统，设备和系统的自然语言规范的软件建模。 解决方案：提取用于从系统的自然语言规范中表示一个或多个句法主体的句法结构，并且对于预定一组和系统的语法对象的各个元模型元素之间的第一映射 生成建模元模型语言模型。 通过跟踪每个识别的句法主体的映射来生成一个或多个元模型元素，并且生成的元模型结构元素被提供用于转换为系统的模型。 版权所有（C）2011，JPO＆INPIT

公开(公告)号：GB2325110B

公开(公告)日：2002-10-16

申请号：GB9709185

申请日：1997-05-06

Applicant: IBM

Inventor： PICKERING JOHN BRIAN ,  STEWART HOYT MCCLURE ,  THOMAS MARK GEORGE STEWART REI ,  HUGHES JEREMY PETER JAMES ,  HULSE BRIAN ,  MAYNARD CAROLINE EDITH

IPC: H04M3/493 ,  H04M3/50

Abstract: A voice processing system comprising a digital trunk processor 135 and a system unit 145 incorporates a local voice recognition resource 185 provided on a digital signal processing board 180, and can access a remote voice recognition resource 320 in a server 300 via a local area network 250. The system provides a barge-in facility for the recognition resource on the remote server, whereby a prompt is played out to the user, and the incoming telephony signal is fed into a voice activity detector on the digital trunk processor. Responsive to a detection of incoming voice activity, the outgoing prompt is terminated, and the incoming data is transferred over the local area network to the remote server for recognition.

公开(公告)号：GB2325112A

公开(公告)日：1998-11-11

申请号：GB9709187

申请日：1997-05-06

Applicant: IBM

Inventor： BUTLER NICHOLAS DAVID ,  HUGHES JEREMY PETER JAMES ,  HULSE BRIAN ,  JORDAN ROBERT MICHAEL ,  MAYNARD CAROLINE EDITH ,  PICKERING JOHN BRIAN ,  RITCHIE ANDREW

IPC: H04M3/493 ,  H04M7/12 ,  H04M3/50

Abstract: A voice processing system comprising a digital trunk processor 135 and a system unit 145 incorporates a local voice recognition resource 185 provided on a digital signal processing board 180, and can access a remote voice recognition resource 320 in a server 300 via a local area network 250. The digital signal processing board 180 is connected to trunk processor 135 via a time division multiplex (TDM) bus 165 and trunk adaptors 160. An application program is provided for controlling the processing of telephone calls and an application interface renders the location of said voice recognition resource 185 transparent to the application program.

公开(公告)号：GB2424742A

公开(公告)日：2006-10-04

申请号：GB0506528

申请日：2005-03-31

Applicant: IBM

Inventor： DOW BARRY NEIL ,  LAWRENCE STEPHEN GRAHAM COPING ,  PICKERING JOHN BRIAN

IPC: G10L15/14 ,  G10L15/00 ,  G10L15/065 ,  G10L15/187

Abstract: A system for automatic recognition of foreign speakers analyses (402) a speech sample and attempts to match the resulting signal to one of a plurality of predetermined sounds to produce recognition, it is determined whether a gap exists in a table of predetermined sounds in a predetermined language and, if a gap exists, a sound from a position in the table near the gap is substituted. This automatic substitution of known foreign pronunciation characteristics improves and modifies the output from standard, monolingual automatic speech recognition (ASR). There is no requirement to implement adaptation or to impose additional manual modification of the baseform pool on the application developer. In addition, the danger of increased ambiguity is removed by introducing an explicit accent and/or language checker (408) which queries known differences across accents.

公开(公告)号：DE69431939T2

公开(公告)日：2003-10-02

申请号：DE69431939

申请日：1994-02-21

Applicant: IBM

Inventor： MINAKAMI MICHAEL KEN ,  HULSE BRIAN ,  COOK JONATHAN ,  PICKERING JOHN BRIAN

IPC: G06F3/16 ,  G06F13/00 ,  H04M3/00 ,  H04M3/493 ,  H04M3/50 ,  H04M3/51 ,  H04M3/527 ,  H04M7/00 ,  H04Q3/00 ,  H04Q3/545

Abstract: A first computer workstation 10 includes a voice response unit for interfacing to a telephone network. The first computer workstation is attached by a communications link 18 to a second computer workstation 20, which includes a server to perform a particular voice processing function, such as text to speech, voice recognition, FAX-back, and so on. For inbound applications the first computer workstation forwards the incoming signal over the communications link to the server on the second computer workstation for real-time processing, whilst for outbound applications, the reverse process occurs.

公开(公告)号：GB2352933A

公开(公告)日：2001-02-07

申请号：GB9917931

申请日：1999-07-31

Applicant: IBM

Inventor： CALDER GARY ,  CLELLAND GEORGE ,  FARRELL ANTHONY ,  MANN ROBERT ,  PICKERING JOHN BRIAN ,  REILLY PAUL

IPC: H04M3/493 ,  H04Q1/45

Abstract: A client device such as a laptop, PDA or mobile phone includes means for converting speech input by a user into DTMF tones which are transmitted over a voice channel to an interactive voice response (IVR) telephony application. For example the application may provide access to a users bank account details and require a tone input as part of the interactive dialogue with the user. The speech may be converted to text before conversion into DTMF tones. An allowed response feature may also be provided which converts the users response to a known valid response of the IVR application. A language conversion feature may further be provided to allow a person to speak to an IVR application which operates in a different language to that of the user. The speech may also be converted to ASCII text or WAP format.

公开(公告)号：GB2409087A

公开(公告)日：2005-06-15

申请号：GB0328793

申请日：2003-12-12

Applicant: IBM

Inventor： CLELLAND GEORGE ,  PICKERING JOHN BRIAN

IPC: G10L15/00 ,  G10L15/22

Abstract: A method in an interactive voice response system comprising: receiving an input audio stream over a voice channel from a user; performing keyword recognition on received input audio as subsequent input audio is being received; and prompting the user with an acknowledgement of the keyword or keywords as subsequent input audio is being received. Preferably, the volume of the speech input is continuously monitored, and recognition results and associated confidence values are combined to select different confirmatory prompts, and the volume is tailored to be the same as, louder than or quieter than the volume of the speech input, so that different types of confirmation can be automatically generated to produce a natural speech-enabled interface.

公开(公告)号：GB2305288A

公开(公告)日：1997-04-02

申请号：GB9518949

申请日：1995-09-15

Applicant: IBM

Inventor： INTERNATIONAL BUSINESS MACHINE ,  COBBETT MICHAEL ,  PICKERING JOHN BRIAN

IPC: G10L15/06 ,  G10L15/14 ,  G10L9/00

Abstract: A speech recognition system for discrete words uses a Hidden Markov Model (HMM), which is nominally adapted to recognise N different isolated words, but which is trained to recognise M different words, where M>N. This is achieved by providing M sets of audio recordings, each set comprising multiple recordings of a respective one of said M words being spoken. Only N different labels are assigned to the M sets of audio recordings, so that at least one of the N labels has two or more sets of audio recordings assigned thereto. These two or more sets of audio recordings correspond to phonetically dissimilar words. The HMM is then trained by inputting each set of audio recordings and its assigned label. The HMM can effectively compensate for the phonetic variations between the different words assigned the same label, thereby avoiding the need to utilise a larger model (ie to use M labels).

公开(公告)号：GB2280820A

公开(公告)日：1995-02-08

申请号：GB9315695

申请日：1993-07-29

Applicant: IBM

Inventor： MINAKAMI MICHAEL KEN ,  HULSE BRIAN ,  COOK JONATHAN ,  PICKERING JOHN BRIAN

IPC: G06F3/16 ,  G06F13/00 ,  H04M3/00 ,  H04M3/493 ,  H04M3/50 ,  H04M3/51 ,  H04M3/527 ,  H04M7/00 ,  H04Q3/00 ,  H04Q3/545

Abstract: A first computer workstation 10 includes a voice response unit for interfacing to a telephone network. The first computer workstation is attached by a communications link 18 to a second computer workstation 20, which includes a server to perform a particular voice processing function, such as text to speech, voice recognition, FAX-back, and so on. For inbound applications the first computer workstation forwards the incoming signal over the communications link to the server on the second computer workstation for real-time processing, whilst for outbound applications, the reverse process occurs.

公开(公告)号：GB2426368A

公开(公告)日：2006-11-22

申请号：GB0510451

申请日：2005-05-21

Applicant: IBM

Inventor： DOYLE JOHN ,  PICKERING JOHN BRIAN

IPC: G10L15/20 ,  G10L15/08 ,  G10L15/14

Abstract: The quality of an input signal into a speech recognition system (310) is measured and a rejection threshold (322) of the speech recognition system (310) is varied in dependence on the input signal quality. If the input signal quality is low, the rejection threshold is reduced and, if the input signal quality is high, the rejection threshold is increased. The measurement of the input signal quality may be based on one or more of the signal-to-noise ratio, loudness, including clipping, and speech signal duration.