公开(公告)号：EP2901445B1

公开(公告)日：2018-01-17

申请号：EP13765743.3

申请日：2013-09-20

Applicant: Cirrus Logic International Semiconductor Ltd.

Inventor： BUERA RODRIGUEZ, Luis ,  VAQUERO AVILES-CASCO, Carlos ,  GARCIA GOMAR, Marta ,  ARTIAGA, Antonio Miguel

IPC: G10L17/04 ,  G10L17/10 ,  G10L17/16 ,  G10L17/24

CPC classification number: G10L17/16 ,  G10L17/04 ,  G10L17/10 ,  G10L17/14 ,  G10L17/24 ,  G10L2015/025

Abstract: Method for text-dependent Speaker Recognition using a speaker adapted Universal Background Model; wherein the speaker adapted Universal Background Model is a speaker adapted Hidden Markov Model comprising channel correction.

公开(公告)号：EP3264333A1

公开(公告)日：2018-01-03

申请号：EP17187085.0

申请日：2013-02-22

Applicant: Samsung Electronics Co., Ltd.

Inventor： KANG, Sangki ,  KIM, Kyungtae ,  LEE, Sanghoon ,  KIM, Yangwook ,  LEE, Chulhwan ,  JUNG, Seokyeong

IPC: G06K9/00 ,  G06F21/32 ,  G10L17/10 ,  G06K9/62 ,  G10L15/25

CPC classification number: G06F21/32 ,  G06K9/00335 ,  G06K9/00892 ,  G06K9/6293 ,  G10L15/25 ,  G10L17/10

Abstract: A system and method is provided that authenticates a user using hybrid biometrics information, such as a user's image information, a user's voice information, etc. The user authentication method includes: acquiring a number of biometrics information; generating a number of authentication information corresponding to the acquired biometrics information; and performing an integral user authentication based on the by generated authentication information.

Abstract translation: 提供了一种使用混合生物特征信息（诸如用户的图像信息，用户的语音信息等）来认证用户的系统和方法。用户认证方法包括：获取多个生物特征信息; 产生对应于所获取的生物测量信息的多个认证信息; 以及基于所生成的认证信息执行完整的用户认证。

公开(公告)号：EP3117428A1

公开(公告)日：2017-01-18

申请号：EP15781848.5

申请日：2015-09-08

Applicant: Frequentis AG

Inventor： KLOPF, Markus

IPC: G10L15/20 ,  G10L17/10 ,  G08G5/00

CPC classification number: G10L15/20 ,  G08G3/00 ,  G08G5/0013 ,  G08G5/0021 ,  G08G5/0056 ,  G10L17/10

Abstract: The invention relates to a method and to a radiotelephony system for identifying and verifying radiotelephony messages (M
1 ...M
3 ) and for associating radiotelephony messages (M
1 ...M
3 ) with vehicles (F
1 ...F
3 ). An announcer specifies the identification (K) at a predetermined point of each radiotelephony message (M
1 ...M
3 ). According to the invention, a) a number of delivered radiotelephony messages (M
1 ...M
3 ) are recorded. Each identification (K) containing the radiotelephony message (M
1 ...M
3 ) is transformed by voice recognition (0) into a digital identification (K
d ). Radiotelephony messages (M
1 ...M
3 ) are extracted, each one being associated with the same digital identification, a biometric data set (B
1 ...B
3 ) is extracted, and said biometric data set (B
1 ...B
3 ) is associated with the respective digital identification (K
d ), and b) then an additional radiotelephony message (M
4 ) is recorded. From the additional radiotelephony message (M
4 ), an additional biometric data set (B
4 ) is extracted. From the recorded biometric data sets (B
1 ...B
3 )), the biometric-dataset (B
1 ) which best matches the additional biometric data set (B
4 ) and the radiotelephony message (M
4 ) of the vehicle (F
1 ) is searched for and the associated identification (K
d ) is associated with said biometric data set (B
1 ).

Abstract translation: 本发明涉及用于识别和验证无线电话消息（M1 ... M3）并将无线电话消息（M1 ... M3）与车辆（F1 ... F3）相关联的方法和无线电话系统。 播音员指定每个无线电话消息（M1 ... M3）的预定点处的标识（K）。 根据本发明，a）记录多个递送的无线电话消息（M1 ... M3）。 包含无线电话消息（M1 ... M3）的每个标识（K）被语音识别（0）转换成数字标识（Kd）。 提取无线电话消息（M1 ... M3），每一个与相同的数字标识相关联，提取生物特征数据集（B1 ... B3），并且所述生物特征数据集（B1 ... B3）相关联 使用相应的数字标识（Kd）和b）然后记录附加的无线电话消息（M4）。 从额外的无线电话消息（M4）中，提取附加的生物测定数据集（B4）。 从记录的生物统计数据集（B1 ... B3）中搜索与车辆（F1）的附加生物统计数据组（B4）和无线电话消息（M4）最匹配的生物体数据集（B1） 相关联的识别（Kd）与所述生物特征数据集（B1）相关联。

公开(公告)号：EP2504745B1

公开(公告)日：2016-08-17

申请号：EP10833498.8

申请日：2010-11-09

Applicant: Samsung Electronics Co., Ltd.

Inventor： KIM, Nam-Hoon ,  CHO, Jeong-Mi ,  PARK, Chi-Youn ,  KIM, Jeong-Su

IPC: G10L15/22 ,  G10L17/10

CPC classification number: G10L15/22 ,  G10L17/10 ,  G10L2015/223 ,  G10L2015/227

Abstract: A communication interface apparatus for a system and a plurality of users is provided. The communication interface apparatus for the system and the plurality of users includes a first process unit configured to receive voice information and face information from at least one user, and determine whether the received voice information is voice information of at least one registered user based on user models corresponding to the respective received voice information and face information; a second process unit configured to receive the face information, and determine whether the at least one user's attention is on the system based on the received face information; and a third process unit configured to receive the voice information, analyze the received voice information, and determine whether the received voice information is substantially meaningful to the system based on a dialog model that represents conversation flow on a situation basis.

公开(公告)号：EP2504745A4

公开(公告)日：2014-12-10

申请号：EP10833498

申请日：2010-11-09

Applicant: SAMSUNG ELECTRONICS CO LTD

Inventor： KIM NAM-HOON ,  CHO JEONG-MI ,  PARK CHI-YOUN ,  KIM JEONG-SU

IPC: G10L17/10 ,  G10L25/78

CPC classification number: G10L15/22 ,  G10L17/10 ,  G10L2015/223 ,  G10L2015/227

Abstract: A communication interface apparatus for a system and a plurality of users is provided. The communication interface apparatus for the system and the plurality of users includes a first process unit configured to receive voice information and face information from at least one user, and determine whether the received voice information is voice information of at least one registered user based on user models corresponding to the respective received voice information and face information; a second process unit configured to receive the face information, and determine whether the at least one user's attention is on the system based on the received face information; and a third process unit configured to receive the voice information, analyze the received voice information, and determine whether the received voice information is substantially meaningful to the system based on a dialog model that represents conversation flow on a situation basis.

Abstract translation: 提供了一种用于系统和多个用户的通信接口装置。 用于系统和多个用户的通信接口装置包括：第一处理单元，被配置为从至少一个用户接收语音信息和面部信息，并且基于用户确定所接收的语音信息是否是至少一个注册用户的语音信息 对应于相应接收的语音信息和面部信息的模型; 第二处理单元，被配置为接收面部信息，并且基于所接收的脸部信息来确定至少一个用户的注意力是否在系统上; 以及第三处理单元，被配置为接收语音信息，分析所接收的语音信息，并且基于在情况基础上表示对话流的对话模型来确定所接收的语音信息对系统是否基本有意义。

公开(公告)号：EP2361430B1

公开(公告)日：2014-06-25

申请号：EP09828719.6

申请日：2009-11-26

Applicant: Persay Ltd.

Inventor： ALEY-RAZ, Almog ,  KRAUSE, Nir Moshe ,  SALMON, Michael Itzhak ,  GAZIT, Ran Yehoshua

IPC: G10L17/24 ,  G10L17/10

CPC classification number: G10L17/24 ,  G10L17/04 ,  G10L17/10

公开(公告)号：EP2713367A1

公开(公告)日：2014-04-02

申请号：EP12006775.6

申请日：2012-09-28

Applicant: Agnitio S.L.

Inventor： Buera Rodriguez, Luis ,  Vaquero Avilés-Casco, Carlos ,  Carcia Gomar, Marta ,  Artiaga, Antonio Miguel

IPC: G10L17/24 ,  G10L17/04 ,  G10L17/16 ,  G10L17/10

CPC classification number: G10L17/16 ,  G10L17/04 ,  G10L17/10 ,  G10L17/14 ,  G10L17/24 ,  G10L2015/025

Abstract: Method for text-dependent Speaker Recognition using a speaker adapted Universal Background Model; wherein the speaker adapted Universal Background Model is a speaker adapted Hidden Markov Model comprising channel correction.

Abstract translation: 使用扬声器适应通用背景模型的文本相关扬声器识别方法; 其中所述扬声器适配的通用背景模型是包括信道校正的扬声器适配的隐马尔可夫模型。

公开(公告)号：EP2028647A1

公开(公告)日：2009-02-25

申请号：EP07114958.7

申请日：2007-08-24

Applicant: Deutsche Telekom AG

Inventor： Runge, Fred ,  Burkhardt, Felix ,  Stegmann, Joachim ,  Müller, Christian

IPC: G10L17/00 ,  G10L15/08

CPC classification number: G10L17/10

Abstract: Verfahren zur automatischen Sprecherklassifizierung, durch ein digitales System, wobei mindestens zwei unterschiedliche Sprecherklassifikationsverfahren auf digitale Sprachdaten angewandt werden und deren Ergebnisse miteinander kombiniert werden.

Abstract translation: 该方法涉及对数字语音数据实现两种不同的说话者分类处理，并将说话者分类处理的结果相互组合。 作为两个扬声器分类处理之一，应用基于梅尔频率 - 阶跃系数（MFCC）的高斯混合模型（GMM），并且将支持向量机（SVM）应用于另一扬声器分类处理，其中， 使用的混合物达到96，并且说话者分类处理的组合在评分水平上进行。 自动扬声器分类的设备还包括独立声明。

公开(公告)号：EP0870300B1

公开(公告)日：2006-04-19

申请号：EP96921329.7

申请日：1996-06-06

Applicant: RUTGERS UNIVERSITY

Inventor： MAMMONE, Richard, J. ,  FARRELL, Kevin ,  SHARMA, Manish ,  DEVANG, Naik ,  ZHANG, Xiaoyu ,  ASSALEH, Khaled ,  LIOU, Han-Sheng

IPC: G10L17/00

CPC classification number: G10L17/10 ,  G10L17/04 ,  G10L17/14 ,  G10L17/18 ,  G10L17/20 ,  G10L25/03

Abstract: The present invention relates to a pattern recognition system (Fig. 1) which uses data fusion to combine data from a plurality of extracted features (60, 61, 62) and a plurality of classifiers (70, 71, 72). Speaker patterns can be accurately verified with the combination of discriminant based and distortion based classifiers. A novel approach using a training set of a "leave one out" data can be used for training the system with a reduced data set (Figs. 7A, 7B, 7C). Extracted features can be improved with a pole filtered method for reducing channel effects (Fig. 11B) and an affine transformation for improving the correlation between training and testing data (Fig. 14).

公开(公告)号：EP4407438A2

公开(公告)日：2024-07-31

申请号：EP24182951.4

申请日：2020-12-15

Applicant: Google LLC

Inventor： SHARIFI, Matthew ,  CARBUNE, Victor

IPC: G06F3/16

CPC classification number: G10L15/22 ,  G10L2015/08820130101 ,  G10L2015/22720130101 ,  G10L17/02 ,  G10L17/18 ,  G10L17/10 ,  G10L17/14 ,  G06F3/167 ,  G10L17/04

Abstract: Implementations relate to automatic generation of speaker features for each of one or more particular text-dependent speaker verifications (TD-SVs) for a user. Implementations can generate speaker features for a particular TD-SV using instances of audio data that each capture a corresponding spoken utterance of the user during normal non-enrollment interactions with an automated assistant via one or more respective assistant devices. For example, a portion of an instance of audio data can be used in response to: (a) determining that recognized term(s) for the spoken utterance captured by that the portion correspond to the particular TD-SV; and (b) determining that an authentication measure, for the user and for the spoken utterance, satisfies a threshold. Implementations additionally or alternatively relate to utilization of speaker features, for each of one or more particular TD-SVs for a user, in determining whether to authenticate a spoken utterance for the user.