公开(公告)号：US20140205116A1

公开(公告)日：2014-07-24

申请号：US13997696

申请日：2012-03-31

Applicant: Charles C. Smith ,  Michelle X. Gong ,  Jennifer A. Healey

Inventor： Charles C. Smith ,  Michelle X. Gong ,  Jennifer A. Healey

IPC: H04R3/00

CPC classification number: H04R3/005 ,  G10L17/00 ,  G10L25/51 ,  H04M3/568 ,  H04M3/569 ,  H04M2201/41 ,  H04M2203/509

Abstract: A system, device, and method for generating an audio output includes a master computing device and a plurality of client computing devices. Each client computing device includes a microphone to record audio signals. The client computing devices generate audio data based on the audio signals and transmit the audio data to the master computing device. The master computing device generates a final, higher quality audio output as a function of the audio data received from collection of participating the client computing devices.

Abstract translation: 用于生成音频输出的系统，设备和方法包括主计算设备和多个客户端计算设备。 每个客户端计算设备包括用于记录音频信号的麦克风。 客户计算设备基于音频信号生成音频数据，并将音频数据传送到主计算设备。 主计算设备根据从参与客户端计算设备的收集接收的音频数据产生最终更高质量的音频输出。

公开(公告)号：US08750489B2

公开(公告)日：2014-06-10

申请号：US12257037

申请日：2008-10-23

Applicant: Youngja Park

Inventor： Youngja Park

IPC: H04M3/00 ,  H04M5/00 ,  H04M1/64 ,  G10L15/00 ,  G10L17/00 ,  G10L15/04

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：US08724779B2

公开(公告)日：2014-05-13

申请号：US13424989

申请日：2012-03-20

Applicant: Rahul P. Akolkar ,  Joao P. Goncalves ,  Jacquelyn A. Martino ,  Mathew S. Thoennes

Inventor： Rahul P. Akolkar ,  Joao P. Goncalves ,  Jacquelyn A. Martino ,  Mathew S. Thoennes

IPC: H04M1/64 ,  H04M1/56

CPC classification number: H04M3/5166 ,  H04M2201/41 ,  H04M2203/6054

Abstract: A small baseline audio sample is sampled when a person initially calls in and the sample is held only for the duration of the call. For each subsequent transfer, a comparison is made to the baseline established from the initial call and at the end of the call the voice sample is discarded so no resources need to be maintained. Speaker verification and VOIP technologies are used to persist the customer's verification information as service representative hand-offs occur.

Abstract translation: 一个小的基线音频样本是在一个人最初呼叫的时候进行采样，而样本只在呼叫期间被保留。 对于每个后续传输，对从初始呼叫建立的基线进行比较，并且在呼叫结束时，丢弃语音样本，因此不需要维护资源。 扬声器验证和VOIP技术用于维护客户的验证信息，因为服务代表切换发生。

公开(公告)号：US20140119520A1

公开(公告)日：2014-05-01

申请号：US14151865

申请日：2014-01-10

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Peeyush JAISWAL ,  Naveen NARAYAN

IPC: H04M3/493 ,  G10L17/00

CPC classification number: G10L17/00 ,  G10L17/005 ,  G10L17/02 ,  G10L17/08 ,  G10L17/14 ,  G10L21/0272 ,  H04M3/42221 ,  H04M3/4936 ,  H04M3/5166 ,  H04M3/5175 ,  H04M2201/41

Abstract: A method and system for using conversational biometrics and speaker identification and/or verification to filter voice streams during mixed mode communication. The method includes receiving an audio stream of a communication between participants. Additionally, the method includes filtering the audio stream of the communication into separate audio streams, one for each of the participants. Each of the separate audio streams contains portions of the communication attributable to a respective participant. Furthermore, the method includes outputting the separate audio streams to a storage system.

Abstract translation: 一种用于在混合模式通信期间使用对话生物识别和扬声器识别和/或验证来过滤语音流的方法和系统。 该方法包括接收参与者之间的通信的音频流。 此外，该方法包括将通信的音频流过滤成单独的音频流，每个参与者一个。 每个单独的音频流包含可归因于相应参与者的通信的部分。 此外，该方法包括将单独的音频流输出到存储系统。

公开(公告)号：US20140088965A1

公开(公告)日：2014-03-27

申请号：US13628486

申请日：2012-09-27

Applicant: POLARIS WIRELESS, INC.

Inventor： Narender Goel

IPC: G10L17/00

CPC classification number: G10L17/00 ,  G10L17/08 ,  H04M3/2281 ,  H04M3/42348 ,  H04M2201/41 ,  H04M2203/6045 ,  H04M2203/6054 ,  H04M2207/18

Abstract: Methods and systems populate a speech signature database with unique speech signatures that are associated with one or more speaker identities and are further associated with one or more mobile stations and/or telephone numbers. Real-time voice signals are compared to the speech signatures in the speech signature database. When a match is found, the mobile station from which the voice signal originated is located in real-time. Further, the associations in the speech signature database are leveraged to find other relevant mobile stations or users and to generate additional associations and to also locate associated users and mobile stations.

Abstract translation: 方法和系统填充具有与一个或多个扬声器身份相关联的独特语音签名的语音签名数据库，并进一步与一个或多个移动站和/或电话号码相关联。 将实时语音信号与语音签名数据库中的语音签名进行比较。 当发现匹配时，发起语音信号的移动台位于实时。 此外，利用语音签名数据库中的关联来查找其他相关的移动站或用户，并产生附加关联并且还定位相关联的用户和移动站。

公开(公告)号：US08660251B2

公开(公告)日：2014-02-25

申请号：US13547253

申请日：2012-07-12

Applicant: Steven M. Miller ,  Lisa A. Seacat

Inventor： Steven M. Miller ,  Lisa A. Seacat

IPC: H04M3/56 ,  H04N7/15 ,  H04L12/18 ,  G06F15/16 ,  G10L17/00

CPC classification number: H04M3/42153 ,  H04M3/56 ,  H04M2201/41 ,  H04M2203/5063

Abstract: A method, system and computer program product for alerting a participant when a topic of interest is being discussed and/or a speaker of interest is speaking during a conference call. A participant to a conference call identifies the topics and/or speakers of interest which is stored for future use along with the participant's contact information. When a participant's identified topic of interest is being discussed and/or a participant's identified speaker of interest is speaking during a conference call, the participant will be alerted to that fact, such as via the means specified in the participant's contact information.

Abstract translation: 当正在讨论感兴趣的主题和/或在电话会议期间正在说话时，提醒参与者的方法，系统和计算机程序产品。 电话会议的参与者识别存储以供将来使用的主题和/或发言者以及参与者的联系信息。 当正在讨论参与者确定的感兴趣的主题和/或参与者确定的感兴趣的扬声器在电话会议期间发言时，参与者将通过参与者的联系信息中指定的方式来提醒该事实。

公开(公告)号：US20130336171A1

公开(公告)日：2013-12-19

申请号：US13971657

申请日：2013-08-20

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Marian CROAK ,  Hossein ESLAMBOLCHI

IPC: H04M3/56

CPC classification number: H04M3/56 ,  H04M3/385 ,  H04M2201/41 ,  H04M2207/20

Abstract: A method and apparatus for providing access to teleconference services using voice recognition technology to receive information on packet networks such as Voice over Internet Protocol (VoIP) and Service over Internet Protocol (SoIP) networks are disclosed. In one embodiment, the service provider enables a caller to enter access information for accessing a conference service using at least one natural language response.

公开(公告)号：US08588377B2

公开(公告)日：2013-11-19

申请号：US11681616

申请日：2007-03-02

Applicant: Shmuel Shaffer ,  Labhesh Patel ,  Mukul Jain ,  Sanjeev Kumar

Inventor： Shmuel Shaffer ,  Labhesh Patel ,  Mukul Jain ,  Sanjeev Kumar

IPC: H04M1/64 ,  H04M1/725

CPC classification number: H04M3/5335 ,  H04M3/436 ,  H04M2201/41

Abstract: A method for grouping voice messages includes extracting a voice signature from a voice message and tagging the voice message with an identification associated with the voice signature. The method also includes grouping the voice message based on the identification.

Abstract translation: 用于分组语音消息的方法包括从语音消息中提取语音签名并用与语音签名相关联的标识来标记语音消息。 该方法还包括基于识别对语音消息进行分组。

公开(公告)号：US20130216029A1

公开(公告)日：2013-08-22

申请号：US13877261

申请日：2011-09-27

Applicant: Mark Pawlewski

Inventor： Mark Pawlewski

IPC: H04M3/22

CPC classification number: H04M3/2281 ,  G10L15/26 ,  H04M3/436 ,  H04M2201/41 ,  H04M2203/6027

Abstract: Fraudulent callers that masquerade as legitimate callers in order to discover details of bank accounts or other accounts are an increasing problem. In order to detect possible fraudsters and preventing them from obtaining such details a method and system is proposed that transform the recorded speech of a batch of incoming calls to strings of phonemes or text. Thereafter similar speech patterns, such as distinct similar phrases or wording, in the recorded speech are determined and calls having similar speech patterns, and preferably also similar acoustic properties, are grouped together and identified as being from the same fraudulent caller. Transactions initiated by the fraudulent caller can as a result be stopped and preferably a voiceprint of the fraudulent caller's speech is generated and stored in a database for further use.

Abstract translation: 为了发现银行帐户或其他账户的细节，虚伪的呼叫者伪装成合法的呼叫者是一个越来越多的问题。 为了检测可能的欺诈者并阻止他们获得这样的细节，提出了将一批来电的记录语音转换成音素或文本字符串的方法和系统。 此后，确定记录的语音中类似的语音模式，例如不同的类似的短语或措词，并且具有相似的语音模式，并且优选地也是类似的声学特性的呼叫被分组在一起并被识别为来自相同的欺诈呼叫者。 可以停止由欺诈性呼叫者发起的交易，并且优选地生成欺骗性呼叫者语音的声纹并将其存储在数据库中以供进一步使用。

公开(公告)号：US08503653B2

公开(公告)日：2013-08-06

申请号：US12074276

申请日：2008-03-03

Applicant: Sudhir Raman Ahuja ,  Jingdong Chen ,  Yiteng Arden Huang ,  Dong Liu ,  Qiru Zhou

Inventor： Sudhir Raman Ahuja ,  Jingdong Chen ,  Yiteng Arden Huang ,  Dong Liu ,  Qiru Zhou

IPC: H04M3/42 ,  H04M1/00 ,  H04L12/16 ,  H04Q11/00

CPC classification number: H04M3/569 ,  G10L17/00 ,  G10L2021/02166 ,  H04M2201/41 ,  H04M2203/5072

Abstract: A method and apparatus for performing active speaker selection in teleconferencing applications illustratively comprises a microphone array module, a speaker recognition system, a user interface, and a speech signal selection module. The microphone array module separates the speech signal from each active speaker from those of other active speakers, providing a plurality of individual speaker's speech signals. The speaker recognition system identifies each currently active speaker using conventional speaker recognition/identification techniques. These identities are then transmitted to a remote teleconferencing location for display to remote participants via a user interface. The remote participants may then select one of the identified speakers, and the speech signal selection module then selects for transmission the speech signal associated with the selected identified speaker, thereby enabling the participants at the remote location to listen to the selected speaker and neglect the speech from other active speakers.