US 11,861,521 B2
System and method for identification and verification
Georgios Spithourakis, London (GB); Pawel Franciszek Budzianowski, London (GB); Michal Lis, London (GB); Avishek Mondal, London (GB); Ivan Vulic, London (GB); Nikola Mrksic, London (GB); Eshan Singhal, London (GB); Benjamin Peter Levin, London (GB); Pei-Hao Su, London (GB); and Tsung-Hsien Wen, London (GB)
Assigned to PolyAI Limited, London (GB)
Filed by PolyAI Limited, London (GB)
Filed on Jan. 19, 2022, as Appl. No. 17/578,826.
Claims priority of application No. 20210100900 (GR), filed on Dec. 21, 2021.
Prior Publication US 2023/0196151 A1, Jun. 22, 2023
Int. Cl. G06N 7/02 (2006.01); G10L 17/22 (2013.01)
CPC G06N 7/02 (2013.01) [G10L 17/22 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented dialogue method comprising:
obtaining, by way of an input, input data relating to speech provided by a user;
deriving, from the input data relating to speech provided by a user, one or more hypotheses for each of a plurality of user data fields corresponding to knowledge based authentication factors, wherein deriving the one or more hypotheses comprises:
performing automatic speech recognition on the input data to generate text from the input data; and
determining, from the text generated, one or more proposed values for each of the plurality of user data fields,
wherein the one or more hypotheses comprise the one or more proposed values for each of the plurality of user data fields;
obtaining one or more reference values for each of the plurality of user data fields for each of one or more candidate users;
calculating a score for at least one candidate user of the one or more candidate users, calculating the score comprising:
calculating a plurality of user data field scores comprising, for each of the plurality of user data fields, a respective user data field score, by comparing the one or more hypotheses and the one or more reference values for the candidate user for the respective user data field;
performing one or more fuzzy logic operations on the plurality of user data field scores;
using the score for a candidate user of the one or more candidate users to perform a verification or identification process for the user;
verifying or identifying the user based on the verification or identification process using the score for the candidate user; and
providing, in response to verifying or identifying the user based on the verification or identification process using the score for the candidate user, access to an authenticated conversational system, the providing access to an authenticated conversational system comprising employing an authenticated interactive voice response (IVR) system to interact with the user.