公开(公告)号：EP4010898A1

公开(公告)日：2022-06-15

申请号：EP20829153.4

申请日：2020-12-15

申请人： Google LLC

发明人： SHARIFI, Matthew ,  CARBUNE, Victor

IPC分类号： G10L17/04 ,  G10L17/14 ,  G10L17/02 ,  G10L17/08 ,  G10L17/18 ,  G10L17/10 ,  G10L15/22 ,  G06F3/16 ,  G10L15/08

公开(公告)号：EP3895160A2

公开(公告)日：2021-10-20

申请号：EP19828037.2

申请日：2019-12-03

申请人： Amazon Technologies, Inc.

发明人： DEEPAK, Yavagal

IPC分类号： G10L15/08 ,  G10L17/08 ,  G10L15/32

公开(公告)号：EP3719679B1

公开(公告)日：2021-06-09

申请号：EP19167174.2

申请日：2019-04-03

发明人： Krivokuca Hahn, Vedrana ,  Marcel, Sébastien

IPC分类号： G06F21/32 ,  G06K9/00 ,  G10L17/06 ,  G10L17/08

公开(公告)号：EP3791390A1

公开(公告)日：2021-03-17

申请号：EP19723575.7

申请日：2019-04-30

申请人： Microsoft Technology Licensing, LLC

发明人： KRUPKA, Eyal ,  ZHANG, Shixiong ,  XIAO, Xiong

IPC分类号： G10L17/00 ,  G10L17/04 ,  G10L17/08 ,  G10L15/06 ,  G10L25/51

公开(公告)号：EP3740949A1

公开(公告)日：2020-11-25

申请号：EP18755848.1

申请日：2018-08-24

申请人： Veridas Digital Authentication Solutions, S.L.

发明人： LÓPEZ ESPEJO, Iván ,  PRIETO CALERO, Santiago ,  IRIARTE RUIZ, Ana ,  RONCAL REDÍN, David ,  SÁNCHEZ YOLDI, Miguel Ángel ,  AZANZA LADRÓN, Eduardo

IPC分类号： G10L17/22 ,  G10L17/04 ,  G10L25/60 ,  G10L17/12 ,  G10L17/08

公开(公告)号：EP3373177B1

公开(公告)日：2020-03-18

申请号：EP18154810.8

申请日：2018-02-02

申请人： Daon Holdings Limited

发明人： BLOUET, Raphael

IPC分类号： G06F21/32 ,  G10L17/08 ,  G06K9/00 ,  G10L17/12 ,  G10L17/18 ,  H04L29/06 ,  H04L9/32

公开(公告)号：EP3100261A1

公开(公告)日：2016-12-07

申请号：EP15722860.2

申请日：2015-05-01

申请人： Google Inc.

发明人： FOERSTER, Jakob Nicolaus ,  CASADO, Diego Melendo

IPC分类号： G10L17/08 ,  G10L17/12 ,  G10L17/22 ,  H04M3/38

CPC分类号： G10L17/20 ,  G06F3/167 ,  G10L17/005 ,  G10L17/02 ,  G10L17/04 ,  G10L17/06 ,  G10L17/08 ,  G10L17/12 ,  G10L17/22 ,  G10L17/24 ,  G10L25/84 ,  H04M3/385

摘要： Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a dynamic threshold for speaker verification are disclosed. In one aspect, a method includes the actions of receiving, for each of multiple utterances of a hotword, a data set including at least a speaker verification confidence score, and environmental context data. The actions further include selecting from among the data sets, a subset of the data sets that are associated with a particular environmental context. The actions further include selecting a particular data set from among the subset of data sets based on one or more selection criteria. The actions further include selecting, as a speaker verification threshold for the particular environmental context, the speaker verification confidence score. The actions further include providing the speaker verification threshold for use in performing speaker verification of utterances that are associated with the particular environmental context.