公开(公告)号：US20100104086A1

公开(公告)日：2010-04-29

申请号：US12257037

申请日：2008-10-23

Applicant: Youngja Park

Inventor： Youngja Park

IPC: H04M3/00

CPC classification number: H04M3/51 ,  G10L15/04 ,  H04M3/42221 ,  H04M2201/40 ,  H04M2201/41 ,  H04M2203/357

Abstract: A system and method for automatic call segmentation including steps and means for automatically detecting boundaries between utterances in the call transcripts; automatically classifying utterances into target call sections; automatically partitioning the call transcript into call segments; and outputting a segmented call transcript. A training method and apparatus for training the system to perform automatic call segmentation includes steps and means for providing at least one training transcript with annotated call sections; normalizing the at least one training transcript; and performing statistical analysis on the at least one training transcript.

Abstract translation: 一种用于自动呼叫分段的系统和方法，包括用于自动检测呼叫转录中的话语之间的边界的步骤和装置; 自动将话语分类为目标通话部分; 自动将通话记录分成通话段; 并输出分段呼叫转录。 用于训练系统执行自动呼叫分段的训练方法和装置包括用于提供至少一个具有注释呼叫部分的训练成绩单的步骤和装置; 规范化至少一个训练成绩单; 以及对所述至少一个训练成绩单执行统计分析。

公开(公告)号：US20090268882A1

公开(公告)日：2009-10-29

申请号：US12428243

申请日：2009-04-22

Applicant: Yun-Wen Lee ,  Liang-Sheng Huang ,  Jia-Lin Shen

Inventor： Yun-Wen Lee ,  Liang-Sheng Huang ,  Jia-Lin Shen

IPC: H04M1/64 ,  G10L15/00

CPC classification number: H04M3/53391 ,  G10L15/26 ,  G10L17/00 ,  G10L2015/226 ,  H04M3/5158 ,  H04M2201/41 ,  H04M2201/42 ,  H04M2203/6045

Abstract: A dialogue system and a speech dialogue operation method are provided. In the system, an inquiry speech signal is generated according to data for specific person and is transferred to a receiver. In response to the inquiry speech signal, a reply speech signal is generated by the receiver, and after the reply speech signal is recognized, a speech correspondence signal is obtained. A ID recognition procedure is performed according to the speech correspondence signal, and a dialogue control signal is generated according to the result of the ID recognition procedure, and accordingly, a response speech signal is generated and transferred to the receiver to continue or terminate the dialogue. When the receiver is recognized to be matched with the specific person, a message delivering procedure is performed to deliver a message to be transferred to the specific person to the receiver.

Abstract translation: 提供对话系统和语音对话操作方法。 在系统中，根据特定人物的数据生成查询语音信号，并将其传送到接收机。 响应于询问语音信号，由接收机生成应答语音信号，并且在应答语音信号被识别之后，获得语音对应信号。 根据语音对应信号进行ID识别程序，根据ID识别程序的结果生成对话控制信号，生成响应语音信号并将其传送到接收机继续或终止对话 。 当接收器被识别为与特定人物匹配时，执行消息传递过程以将要传送到特定人的消息传送到接收者。

公开(公告)号：US20090252304A1

公开(公告)日：2009-10-08

申请号：US12154778

申请日：2008-05-27

Applicant: Timothy A. Timmins ,  John S. Miller ,  Nathan B. Baker ,  Christopher A. Huey ,  Philip A. Ljubicich

Inventor： Timothy A. Timmins ,  John S. Miller ,  Nathan B. Baker ,  Christopher A. Huey ,  Philip A. Ljubicich

IPC: H04M1/64

CPC classification number: H04M3/4878 ,  H04M3/42068 ,  H04M3/4931 ,  H04M2201/41

Abstract: In communications systems where multiple users may be associated with the same account, the voiceprint of a user is used to identify a profile of the user, facilitating identification of the user's preferences without requesting additional information from the user. The system may also monitor messages played to individual users, to control the playing of the messages to users. Calls made by users associated with the same account may be allocated to the caller initiating the call in bills, based on voiceprints. Voiceprints may be collected for use in deriving voiceprint samples during registration of a user with the system and during conversations between the user and the system. A voiceprint comparison threshold may be set based on the type of phone calling the system, to compensate for noise. Changes in phone numbers and/or addresses of users are monitored and the threshold is adjusted if changes are found.

Abstract translation: 在多个用户可能与同一帐户相关联的通信系统中，用户的声纹用于识别用户的简档，便于识别用户的偏好而不需要用户的附加信息。 该系统还可以监视对各个用户播放的消息，以控制向用户播放消息。 基于声波纹的用户可以将与同一帐户相关联的用户的呼叫分配给呼叫者发起账单。 可以收集Voiceprint用于在用户与系统注册期间以及在用户与系统之间的对话期间导出声纹样本。 可以基于呼叫系统的电话的类型来设置声纹比较阈值，以补偿噪声。 监控用户的电话号码和/或地址的更改，如果发现更改，则调整阈值。

公开(公告)号：US07590232B2

公开(公告)日：2009-09-15

申请号：US11091129

申请日：2005-03-28

Applicant: John A. Carter ,  Ronald D. DeBerry ,  Kurt Schuder

Inventor： John A. Carter ,  Ronald D. DeBerry ,  Kurt Schuder

IPC: H04M3/58

CPC classification number: H04M3/2281 ,  G07C9/00158 ,  G10L17/00 ,  H04M2201/41

Abstract: A client identity verification system and method which includes receiving a spoken voice sample from a client. Identifying the match of the client, verifying client information, changing parameters in a conditional database and communicating alerts based on the results. The method also includes matching the received spoken voice sample against a stored voiceprint and voiceprint template of random phrases. The method then also includes verifying the phone number of the client.

Abstract translation: 客户身份验证系统和方法，其包括从客户端接收口语语音样本。 识别客户端的匹配，验证客户端信息，更改条件数据库中的参数以及根据结果传达警报。 该方法还包括将接收到的语音语音样本与存储的声纹和随机短语的声纹模板进行匹配。 该方法还包括验证客户端的电话号码。

公开(公告)号：US20090190737A1

公开(公告)日：2009-07-30

申请号：US12022855

申请日：2008-01-30

Applicant: David L. Dunmire ,  Jeff P. Clark ,  Brian M. Novack

Inventor： David L. Dunmire ,  Jeff P. Clark ,  Brian M. Novack

IPC: H04M3/42

CPC classification number: H04M3/42059 ,  G10L17/02 ,  H04L63/0861 ,  H04M2201/405 ,  H04M2201/41 ,  H04M2203/60

Abstract: A disclosed method of authenticating a telephone caller includes receiving an authentication request for the telephone caller that includes a voice recording of the caller and household information indicative of a household, creating a first voice pattern based on the voice recording of the caller, retrieving a second voice pattern using at least some of the household information, comparing the two voice patterns, and transmitting information indicative of the results of the comparison. The transmitted information may include information that the telephone caller is authenticated, is not authenticated, or that the comparison results were inconclusive. The method may include monitoring telephone calls with the household, and if it is determined that a voice on at least some of the telephone calls is associated with the household, creating the second voice pattern based on the voice, and associating the second voice pattern to the household. The method may further include disassociating the second voice pattern from the household if it is determined that a change of household has occurred.

Abstract translation: 公开的认证电话呼叫者的方法包括：接收对包括呼叫者的语音记录和指示家庭的家庭信息的电话呼叫者的认证请求，基于呼叫者的语音记录创建第一语音模式，检索第二个 使用至少一些家庭信息的语音模式，比较两种语音模式，以及发送指示比较结果的信息。 所发送的信息可以包括电话呼叫者被认证，未认证的信息，或者比较结果不确定。 该方法可以包括监视与家庭的电话呼叫，并且如果确定至少一些电话呼叫中的语音与家庭相关联，则基于语音创建第二语音模式，并将第二语音模式与 家庭。 如果确定已经发生家庭变更，该方法还可以包括将第二语音模式与家庭分离。

公开(公告)号：US07539290B2

公开(公告)日：2009-05-26

申请号：US11129253

申请日：2005-05-13

Applicant: William Charles Gormley Ortel

Inventor： William Charles Gormley Ortel

IPC: H04M1/64

CPC classification number: H04M3/56 ,  H04M3/567 ,  H04M2201/38 ,  H04M2201/41

Abstract: There is provided a system for facilitating a conference call. The system includes a module to generate a real-time voiceprint from a voice input of a participant in the conference call, and a module to provide information indicative of the participant based on the real-time voiceprint.

Abstract translation: 提供了一种便于电话会议的系统。 该系统包括用于从会议呼叫中的参与者的语音输入生成实时声纹的模块，以及用于基于实时声纹提供指示参与者的信息的模块。

公开(公告)号：US20090046841A1

公开(公告)日：2009-02-19

申请号：US12218145

申请日：2008-07-11

Applicant: Stephen L. Hodge

Inventor： Stephen L. Hodge

IPC: H04M3/00 ,  G10L17/00 ,  H04L9/32

CPC classification number: H04M3/382 ,  A61K31/435 ,  A61K31/44 ,  A61K31/505 ,  G06F17/30256 ,  G06F17/30259 ,  G06F17/30268 ,  G06F17/30528 ,  G06F17/30743 ,  G06F17/30876 ,  G06K7/10366 ,  G06K7/10396 ,  G06K9/00288 ,  G06K9/6212 ,  G06K19/10 ,  H04M1/67 ,  H04M1/72572 ,  H04M3/20 ,  H04M3/2218 ,  H04M3/2281 ,  H04M3/38 ,  H04M3/385 ,  H04M3/42025 ,  H04M3/42221 ,  H04M2201/41 ,  H04M2201/42 ,  H04M2203/305 ,  H04M2203/558 ,  H04M2203/6054 ,  H04N5/33

Abstract: Disclosed is a secure telephone call management system for authenticating users of a telephone system in an institutional facility. Authentication of the users is accomplished by using a personal identification number, preferably in conjunction with speaker independent voice recognition and speaker dependent voice identification. When a user first enters the system, the user speaks his or her name which is used as a sample voice print. During each subsequent use of the system, the user is required to speak his or her name. Voice identification software is used to verify that the provided speech matches the sample voice print. The secure system includes accounting software to limit access based on funds in a user's account or other related limitations. Management software implements widespread or local changes to the system and can modify or set any number of user account parameters.

Abstract translation: 公开了一种用于认证机构设施中的电话系统的用户的安全电话呼叫管理系统。 用户的认证是通过使用个人识别号来实现的，优选地与扬声器独立的语音识别和与扬声器相关的声音识别结合在一起。 当用户首次进入系统时，用户说出用作示例语音打印的他或她的名字。 在系统的后续使用期间，用户必须说出他或她的姓名。 语音识别软件用于验证所提供的语音是否与样本语音打印匹配。 安全系统包括会计软件，以根据用户帐户中的资金或其他相关限制来限制访问。 管理软件对系统进行广泛或本地的更改，并可以修改或设置任意数量的用户帐户参数。

公开(公告)号：US07489766B2

公开(公告)日：2009-02-10

申请号：US11360527

申请日：2006-02-24

Applicant: Sanford J. Morganstein ,  Sergey Zaks

Inventor： Sanford J. Morganstein ,  Sergey Zaks

IPC: H04M1/64

CPC classification number: G10L17/005 ,  G06F21/32 ,  G10L17/00 ,  G10L17/22 ,  H04M3/2281 ,  H04M3/229 ,  H04M3/38 ,  H04M3/382 ,  H04M3/385 ,  H04M3/387 ,  H04M3/42 ,  H04M3/42042 ,  H04M3/42059 ,  H04M2201/40 ,  H04M2201/41 ,  H04M2242/22 ,  H04Q3/72 ,  H04Q2213/13003 ,  H04Q2213/1305 ,  H04Q2213/13091 ,  H04Q2213/13093 ,  H04Q2213/13095 ,  H04Q2213/13103 ,  H04Q2213/13141 ,  H04Q2213/13175 ,  H04Q2213/13204 ,  H04Q2213/13271 ,  H04Q2213/13282 ,  H04Q2213/13353

Abstract: In one embodiment, a system authenticates a caller before providing the caller with access to one or more secured resources. Identification and verification data is stored during system enrollment of known persons. A processing system, in a current call initiated by a caller and prior to identification of the caller, receives voice-based information for the caller including a digital representation of a telephone number spoken by the caller in the current call, the received information including voice-based identification information and voice-based verification information for the caller. The processing system automatically compares the received identification information and the received verification information for the caller with the stored identification and verification data for known persons, respectively, to uniquely identify and verify the identity of the caller as a particular known person. The processing system authenticates the caller for access to the one or more secured resources in response.

Abstract translation: 在一个实施例中，系统在向呼叫者提供对一个或多个安全资源的访问之前认证呼叫者。 识别和验证数据在已知人员的系统注册期间存储。 处理系统在由呼叫者发起的当前呼叫中并且在主叫方识别之前，接收呼叫者的基于语音的信息，包括呼叫者在当前呼叫中所说出的电话号码的数字表示，所接收的信息包括声音 的呼叫者识别信息和基于语音的验证信息。 处理系统分别自动将所接收的识别信息和接收到的呼叫者验证信息与所存储的已知人员的识别和验证数据进行比较，以唯一地识别和验证作为特定已知人员的呼叫者的身份。 处理系统对呼叫者进行认证以响应于访问一个或多个安全资源。

公开(公告)号：US20080273673A1

公开(公告)日：2008-11-06

申请号：US11742660

申请日：2007-05-01

Applicant: Magnus OTTOSSON ,  Mats ORMIN ,  Hanshenric CARENBORN ,  Russel HUGHES ,  Robin PETERSSON

Inventor： Magnus OTTOSSON ,  Mats ORMIN ,  Hanshenric CARENBORN ,  Russel HUGHES ,  Robin PETERSSON

IPC: H04M1/64

CPC classification number: H04M3/436 ,  G10L15/26 ,  H04M1/6505 ,  H04M1/665 ,  H04M3/42034 ,  H04M3/42059 ,  H04M3/42093 ,  H04M2201/41 ,  H04M2203/2005 ,  H04M2242/22

Abstract: A method of handling an incoming call from a first communication device addressed to a second communication device, the first and second communication devices being operative in a communication network, wherein the second communication device or the communication network comprises an answering unit. The answering unit includes a privacy data unit comprising a privacy code message for bypassing the answering unit. The method comprises receiving, by means of the answering unit, the incoming call intended for and addressed to the second communication device. Furthermore, the method comprises determining whether also a code message is received by the answering unit. If so, the method comprises comparing the received code message with the privacy code message stored in the privacy data unit to determine whether the incoming call should bypass the answering unit and, thereby, be directly forwarded to the second communication device being addressed by the first communication device in the incoming call. Thereby, a user of the second communication device may be permitted to intercept the incoming call from the calling first communication device.

Abstract translation: 一种处理来自寻址到第二通信设备的第一通信设备的呼入的方法，所述第一和第二通信设备在通信网络中可操作，其中所述第二通信设备或所述通信网络包括应答单元。 应答单元包括隐私数据单元，其包括用于绕过应答单元的隐私代码消息。 该方法包括通过应答单元接收旨在和寻址到第二通信设备的呼入。 此外，该方法包括确定应答单元是否接收到代码消息。 如果是，则该方法包括将接收到的代码消息与存储在隐私数据单元中的隐私代码消息进行比较，以确定来话呼叫是否应该绕过应答单元，从而直接转发到由第一通信设备寻址的第二通信设备 通话设备在来电。 从而，可以允许第二通信设备的用户拦截来自呼叫第一通信设备的呼入。

公开(公告)号：US20080192901A1

公开(公告)日：2008-08-14

申请号：US11780120

申请日：2007-07-19

Applicant: Marc Mumm ,  Raja Kuppuswamy

Inventor： Marc Mumm ,  Raja Kuppuswamy

IPC: H04M11/00

CPC classification number: H04L63/0861 ,  G10L17/22 ,  H04M3/38 ,  H04M3/385 ,  H04M2201/41 ,  H04W12/06 ,  H04W12/08

Abstract: Digital process for authentication of a user of a telecommunications or data network for access to protected data or a service reserved for a defined circle of users or for the use of data currently entered by the user, wherein a voice sample currently enunciated during an access attempt by the user on a telecommunications or VoIP terminal device is routed to a voice analysis unit and, herein, a current voice profile is computed and this is compared in a voice profile comparison unit against a previously stored initial voice profile and, in response to a positive comparison result, the user is authenticated and a first control signal enabling access, but in response to a negative comparison result a second control signal disabling access or triggering a substitute authentication procedure is generated.

Abstract translation: 用于认证用于访问受保护数据的电信或数据网络的数字处理或为定义的用户圈保留的服务或用于使用当前由用户输入的数据的数字处理，其中当前在接入尝试期间被发送的语音样本 由电信或VoIP终端设备上的用户被路由到语音分析单元，并且在此，计算当前语音配置文件，并且在语音配置文件比较单元中与先前存储的初始语音配置文件进行比较，并且响应于 肯定的比较结果，用户被认证，并且第一控制信号能够访问，但是响应于否定的比较结果，产生禁用访问或触发替代认证过程的第二控制信号。