US 11,783,807 B2
Voice response systems based on personalized vocabulary and user profiling—personalized linguistics AI engines
Shikhar Kwatra, Raleigh, NC (US); Adam Lee Griffin, Dubuque, IA (US); Sarbajit K. Rakshit, Kolkata (IN); and Laura Grace Ellis, Austin, TX (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Jul. 24, 2020, as Appl. No. 16/947,235.
Prior Publication US 2022/0028374 A1, Jan. 27, 2022
Int. Cl. G10L 15/07 (2013.01); G10L 15/22 (2006.01); G10L 15/06 (2013.01); G10L 15/19 (2013.01); G06N 3/02 (2006.01); G10L 15/16 (2006.01)
CPC G10L 15/07 (2013.01) [G06N 3/02 (2013.01); G10L 15/063 (2013.01); G10L 15/16 (2013.01); G10L 15/19 (2013.01); G10L 15/22 (2013.01); G10L 2015/223 (2013.01); G10L 2015/227 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method for personalized voice responses, the method comprising:
gathering a plurality of user data from an Internet of Things (IoT) connected sensor;
identifying a personalized vocabulary based on the gathered plurality of user data;
training a voice response system based on the gathered plurality of user data and the identified personalized vocabulary, using a bi-directional long short term memory (Bi-LSTM) training module which contextually identifies, classifies, and learns using pattern analysis, and a text and image information based convolutional neural network (TI-CNN) module;
receiving a verbal request; and
responding to the received verbal request using the trained voice response system and a similar and closely located user's personalized vocabulary, as necessary, wherein the similar and closely located user is identified using a regional linguistic corpus which employs an ontology for similarity of vocabulary.