公开(公告)号：GB2333877A

公开(公告)日：1999-08-04

申请号：GB9900679

申请日：1999-01-14

Applicant: MOTOROLA INC

Inventor： SRENGER EDWARD ,  MEUNIER JEFFREY A ,  KUSHNER WILLIAM M

IPC: G10L15/06 ,  G10L5/06

Abstract: A standard filler, or garbage model, is calculated for the detection of out of vocabulary utterances. Gobal statistical parameters are calculated (116) based upon the statistical parameters for new training data and a garbage model (118) is updated based upon the global statistical parameters. This is carried out on-line while the user is enrolling the vocabulary. The garbage model is preferably an average speaker model, representative of all the speech data enrolled by the user to date. The garbage model is preferably obtained as a by-product of the vocabulary enrollment procedure and is similar in it characteristics and topology to all the other regular vocabulary HMMs.

公开(公告)号：CA2189249A1

公开(公告)日：1996-10-03

申请号：CA2189249

申请日：1996-01-29

Applicant: MOTOROLA INC

Inventor： KUSHNER WILLIAM M ,  SRENGER EDWARD ,  HARTMAN MATTHEW A

IPC: G10L15/14 ,  G10L9/00

Abstract: In a statistical based speech recognition system, one of the key issues is the selection of the Hidden Markov Model that best matches a given sequence of feature observations. The problem is usually addressed by the calculation of the maximum likelihood, ML, state sequence by means of a Viterbi or other decoder. Noise or inadequate training can produce an ML sequence associated with a Hidden Markov Model other than the correct model. The method of the present invention provides improved robustness by combining the standard ML state sequence score (416) with an additional path core (418) derived from the dynamics of the ML score as a function of time. These two scores, when combined, form a hybrid metric (420) that, when used with the decoder, optimizes selection of the correct Hidden Markov Model (422).

公开(公告)号：CA2601719A1

公开(公告)日：2006-09-28

申请号：CA2601719

申请日：2006-02-21

Applicant: MOTOROLA INC

Inventor： ROKUSEK DANIEL S ,  SRENGER EDWARD ,  WEIRICH KEVIN L

IPC: G06F9/00

公开(公告)号：GB2333877B

公开(公告)日：2001-08-08

申请号：GB9900679

申请日：1999-01-14

Applicant: MOTOROLA INC

Inventor： SRENGER EDWARD ,  MEUNIER JEFFREY ARTHUR ,  KUSHNER WILLIAM M

IPC: G10L15/06 ,  G10L15/14 ,  G10L15/20

Abstract: The present invention provides a method of calculating, within the framework of a speaker dependent system, a standard filler, or garbage model, for the detection of out-of-vocabulary utterances. In particular, the method receives new training data in a speech recognition system (202); calculates statistical parameters for the new training data (204); calculates global statistical parameters based upon the statistical parameters for the new training data (206); and updates a garbage model based upon the global statistical parameters (208). This is carried out on-line while the user is enrolling the vocabulary. The garbage model described in this disclosure is preferably an average speaker model, representative of all the speech data enrolled by the user to date. Also, the garbage model is preferably obtained as a by-product of the vocabulary enrollment procedure and is similar in it characteristics and topology to all the other regular vocabulary HMMs.

公开(公告)号：CA2646340A1

公开(公告)日：2007-09-20

申请号：CA2646340

申请日：2007-03-12

Applicant: MOTOROLA INC

Inventor： ISAAC EMAD S ,  SRENGER EDWARD ,  ROKUSEK DANIEL S

IPC: G10L19/00

Abstract: A method of providing information storage by means of Automatic Speech Recognition through a communication device of a vehicle comprises establishing a voice communication between an external source and a user of the vehicle, receiving information from the external source, processing the received information using an Automatic Speech Recognition unit in the vehicle and storing the recognized speech in textual form for future retrieval or use.

公开(公告)号：AU681058B2

公开(公告)日：1997-08-14

申请号：AU5353196

申请日：1996-01-29

Applicant: MOTOROLA INC

Inventor： HARTMAN MATTHEW A ,  KUSHNER WILLIAM M ,  SRENGER EDWARD

IPC: G10L15/14 ,  G10L9/00

Abstract: In a statistical based speech recognition system, one of the key issues is the selection of the Hidden Markov Model that best matches a given sequence of feature observations. The problem is usually addressed by the calculation of the maximum likelihood, ML, state sequence by means of a Viterbi or other decoder. Noise or inadequate training can produce a ML sequence associated with a Hidden Markov Model other than the correct model. The method of the present invention provides improved robustness by combining the standard ML state sequence score (416) with an additional path score (418) derived from the dynamics of the ML score as a function of time. These two scores, when combined, form a hybrid metric (420) that, when used with the decoder, optimizes selection of the correct Hidden Markov Model (422).

公开(公告)号：AU5353196A

公开(公告)日：1996-10-16

申请号：AU5353196

申请日：1996-01-29

Applicant: MOTOROLA INC

Inventor： HARTMAN MATTHEW A ,  KUSHNER WILLIAM M ,  SRENGER EDWARD

IPC: G10L15/14 ,  G10L9/00

Abstract: In a statistical based speech recognition system, one of the key issues is the selection of the Hidden Markov Model that best matches a given sequence of feature observations. The problem is usually addressed by the calculation of the maximum likelihood, ML, state sequence by means of a Viterbi or other decoder. Noise or inadequate training can produce a ML sequence associated with a Hidden Markov Model other than the correct model. The method of the present invention provides improved robustness by combining the standard ML state sequence score (416) with an additional path score (418) derived from the dynamics of the ML score as a function of time. These two scores, when combined, form a hybrid metric (420) that, when used with the decoder, optimizes selection of the correct Hidden Markov Model (422).

公开(公告)号：WO2007008312A2

公开(公告)日：2007-01-18

申请号：PCT/US2006021968

申请日：2006-06-06

Applicant: MOTOROLA INC ,  ROKUSEK DANIEL S ,  KAMBHAMPATI KRANTI K ,  PIKET JAMES B ,  SRENGER EDWARD

Inventor： ROKUSEK DANIEL S ,  KAMBHAMPATI KRANTI K ,  PIKET JAMES B ,  SRENGER EDWARD

IPC: H04W88/02

CPC classification number: H04W88/02

Abstract: An apparatus and method for enhancing a handheld communication device via a telematics system in a vehicle is disclosed. Audio received at the handheld device is transferred to the telematics system in the vehicle. The received audio is then analyzed to determine whether it contains speech, and if so, an audio present signal is generated and the received audio is recorded into a memory coupled to the telematics system. The user can then engage the user interface of the telematics system to replay the recorded audio. Bluetooth protocol is preferably used to establish a channel between the handheld device and the telematics system, which can occur automatically when the two are in proximity. Analysis of the received audio preferably comprises use of a voice detector as part of a speech recognition system otherwise used by the telematics system to assess spoken commands. The memory is preferably overwritten with the latest audio sent from the handheld device to the telematics system, such that engaging the telematics system for playback of the recorded audio will repeat only the last audio sent.

Abstract translation: 公开了一种用于通过车辆中的远程信息处理系统增强手持通信设备的装置和方法。 在手持设备处接收的音频被传送到车辆中的远程信息处理系统。 然后分析接收到的音频以确定其是否包含语音，如果是，则产生音频呈现信号，并将接收的音频记录到耦合到远程信息处理系统的存储器中。 然后，用户可以接合远程信息处理系统的用户界面来重放记录的音频。 蓝牙协议优选地用于在手持设备和远程信息处理系统之间建立信道，当两者处于邻近状态时，该协议可以自动发生。 对所接收的音频的分析优选地包括使用语音检测器作为语音识别系统的一部分，否则由远程信息处理系统用于评估语音命令。 存储器优选地被从手持设备发送到远程信息处理系统的最新音频被覆盖，使得接合远程信息处理系统用于重放所记录的音频将仅重复发送的最后音频。

公开(公告)号：WO2007106758B1

公开(公告)日：2008-07-31

申请号：PCT/US2007063751

申请日：2007-03-12

Applicant: MOTOROLA INC ,  ISAAC EMAD S ,  ROKUSEK DANIEL S ,  SRENGER EDWARD

Inventor： ISAAC EMAD S ,  ROKUSEK DANIEL S ,  SRENGER EDWARD

IPC: G10L19/00

CPC classification number: G10L15/22

Abstract: A method of providing information storage by means of Automatic Speech Recognition through a communication device of a vehicle comprises establishing a voice communication between an external source and a user of the vehicle, receiving information from the external source, processing the received information using an Automatic Speech Recognition unit in the vehicle and storing the recognized speech in textual form for future retrieval or use.

Abstract translation: 通过车辆的通信装置通过自动语音识别提供信息存储的方法包括在外部源和车辆的用户之间建立语音通信，从外部源接收信息，使用自动语音处理所接收的信息 将识别单元存储在车辆中，并将识别的语音存储在文本形式中，以备将来检索或使用。

公开(公告)号：WO2006101649A2

公开(公告)日：2006-09-28

申请号：PCT/US2006006053

申请日：2006-02-21

Applicant: MOTOROLA INC ,  SRENGER EDWARD ,  ROKUSEK DANIEL S ,  WEIRICH KEVIN L

Inventor： SRENGER EDWARD ,  ROKUSEK DANIEL S ,  WEIRICH KEVIN L

IPC: G06F9/00

CPC classification number: G06F9/4446

Abstract: A method and apparatus for adapting a help menu on a user interface, utilizing an input method such as a speech recognition system, for increased efficiency. A list of menu items is presented on the user interface including an optional menu item to reinstate any previously removed menu items. A user selects an item from the menu, such as a help menu, which can then be removed from the list of menu items in accordance with predetermined criteria. The criteria can include how many times the menu item has been accessed and when. In this way, help menu items that are familiar to a user are removed to provide an abbreviated help menu which is more efficient and less frustrating to a user, particularly in a busy and distracting environment such as a vehicle.

Abstract translation: 一种用于在用户界面上调整帮助菜单的方法和装置，利用诸如语音识别系统的输入方法来提高效率。 在用户界面上显示菜单项的列表，包括可选的菜单项，以恢复任何先前删除的菜单项。 用户从菜单中选择一个项目，例如帮助菜单，然后可以根据预定标准从菜单项目列表中删除。 标准可以包括菜单项被访问次数和时间。 以这种方式，删除用户熟悉的帮助菜单项，以提供简化的帮助菜单，该菜单对于用户而言更有效并且不那么令人沮丧，特别是在诸如车辆的忙碌和令人分心的环境中。