Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
VOICEPRINT IDENTIFICATION METHOD, MODEL TRAINING METHOD AND SERVER
Document Type and Number:
WIPO Patent Application WO/2020/073694
Kind Code:
A1
Abstract:
Provided is a voiceprint identification method, comprising: acquiring target voice information to be identified (101); acquiring target feature information of the target voice information by means of a voiceprint identification model, wherein the voiceprint identification model is obtained by training according to a first loss function and a second loss function, the first loss function falls within normalized exponential functions and the second loss function falls within centralization functions (102); and determining a voiceprint identification result according to the target feature information and registration feature information, wherein the registration feature information is obtained after voice information of an object to be identified passes through the voiceprint identification model (103). Further provided are a model training method and a server. Using two types of functions to monitor and study a voice identification model at the same time can make a depth feature have better distinguishability so as to improve the identification performance.

More Like This:
WO/2022/177224ELECTRONIC DEVICE AND OPERATING METHOD OF ELECTRONIC DEVICE
JP2001255887SPEECH RECOGNITION DEVICE, SPEECH RECOGNITION METHOD AND MEDIUM RECORDED WITH THE METHOD
WO/2022/205249AUDIO FEATURE COMPENSATION METHOD, AUDIO RECOGNITION METHOD, AND RELATED PRODUCT
Inventors:
LI NA (CN)
TUO DEYI (CN)
Application Number:
PCT/CN2019/093792
Publication Date:
April 16, 2020
Filing Date:
June 28, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
TENCENT TECH SHENZHEN CO LTD (CN)
International Classes:
G10L17/04; G10L17/18
Domestic Patent References:
WO2005055200A12005-06-16
Foreign References:
CN107146624A2017-09-08
CN103971690A2014-08-06
CN104157290A2014-11-19
CN102270451A2011-12-07
Other References:
WEN, YANDONG ET AL.: "A Discriminative Feature Learning Approach for Deep Face Recognition", COMPUTER VISION-ECCV 2016: 14TH EUROPEAN CONFERENCE, vol. 9911, no. 558, 14 October 2016 (2016-10-14), pages 499 - 515, XP047355154, ISSN: 0302-9743
See also references of EP 3866163A4
Attorney, Agent or Firm:
KANGXIN PARTNERS, P.C. (CN)
Download PDF:
View/Download PDF      PDF Help



 
Previous Patent: ELECTROLYTIC DEVICE FOR WASHING MACHINE

Next Patent: TFT SUBSTRATE, DISPLAY PANEL AND DISPLAY APPARATUS