US 11,856,129 B2
Systems and methods to manage models for call data
Sara McDole, Midlothian, VA (US); Premkumar Dhayalan, Richmond, VA (US); Thomas Zahorik, Charlottesville, VA (US); Saiteja Kamuni, Richmond, VA (US); Sri Penmatsa, Richmond, VA (US); Vamsi Rallabandi, Richmond, VA (US); Danni Tang, Richmond, VA (US); and Suresh Pothireddy, Richmond, VA (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Jun. 1, 2022, as Appl. No. 17/829,556.
Application 17/829,556 is a continuation of application No. 16/879,933, filed on May 21, 2020, granted, now 11,381,677.
Application 16/879,933 is a continuation of application No. 16/693,673, filed on Nov. 25, 2019, granted, now 10,694,024, issued on Jun. 23, 2020.
Prior Publication US 2022/0294897 A1, Sep. 15, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. H04M 3/22 (2006.01); G06Q 20/30 (2012.01); H04M 3/51 (2006.01)
CPC H04M 3/2218 (2013.01) [G06Q 20/305 (2013.01); H04M 3/5175 (2013.01); H04M 3/5183 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for performing call analysis comprising:
one or more memory units storing instructions; and
one or more processors that execute the instructions to perform operations comprising:
receiving, via a network, an electronically tagged model input representing a call record;
accessing a computerized neural network model comprising utterance-related model parameters, the computerized neural network model comprising (i) a recurrent neural network and (ii) being configured to intake the electronically tagged model input and analyze the electronically tagged model input; and
embedding, via the computerized neural network model, based on the utterance-related model parameters, an electronic indicator at a segment of the electronically tagged model input,
wherein embedding the electronic indicator comprises determining placement of the electronic indicator at the segment based on the utterance-related model parameters, the computerized neural network model applying one or more thresholds associated with the utterance-related model parameters to one or more portions of the electronically tagged model input.