Granted Patent B2

US 12,288,194 · App. 17/448,001 · Granted Apr 29, 2025

Inventors: Vivedha Elango (Chennai, IN); Soundararajan Dhakshinamoorthy (Chennai, IN); Srividya Thyagarajan (Chennai, IN); Ninad D. Sathaye (Bangalore, IN); Gregory J. Boss (Saginaw, MI); Santhosh Kumar Gopynadhan (Chennai, IN)

Assignee: Optum, Inc.

G06Q10/1097G06F21/629G06N5/022G06N5/04G06Q10/063116H04M3/5238G10L15/02G10L15/26

Abstract

Claims (80)

1. A system comprising:

memory; and

one or more processors communicatively coupled to the memory, the one or more processors configured to:

receive call-intent data for a caller during a voice-based interaction using a real-time communication channel with a user computing entity of the caller;

determine an unadjusted time estimate and a first confidence score associated with the unadjusted time estimate using the call-intent data as input to a first machine-learning time model;

in response to the first confidence score satisfying a first threshold, determine a time adjustment and a second confidence score associated with the time adjustment using an identity of a customer service agent assigned to the voice-based interaction of the caller as input to a second machine-learning time model;

in response to the second confidence score satisfying a second threshold, determine a predicted call duration with the customer service agent based at least in part on the unadjusted time estimate and the time adjustment; and

schedule an interaction with the customer service agent to address a call-intent for the caller at least in part by:

accessing a calendar of the caller using a direct communication protocol of an executable application operating on the user computing entity of the caller, wherein the direct communication protocol enables data transfer from the user computing entity of the caller during the interaction;

identifying one or more candidate time-periods within the calendar of the caller having a time-period duration equal to or greater than the predicted call duration with the customer service agent, wherein the time-period duration begins when the interaction with the customer service agent is set to begin;

presenting at least a subset of the one or more candidate time-periods using an interactive user interface displayable on the user computing entity of the caller;

receiving input identifying a selected time-period of the one or more candidate time-periods; and

scheduling the interaction with the customer service agent to address the call-intent for the caller within the selected time-period.

2. The system of claim 1 , wherein to receive the call-intent data for the caller, the one or more processors are further configured to:

receive audio comprising a caller's voice for a call;

execute a voice-to-text conversion of the audio to generate transcription data; and

extract one or more keywords from the transcription data.

3. The system of claim 1 , wherein accessing the calendar of the caller comprises one of:

receiving calendar data from the user computing entity of the caller during the voice-based interaction; or

accessing a web-accessible calendar of the caller.

4. The system of claim 1 , wherein identifying the one or more candidate time-periods within the calendar of the caller for the interaction having the predicted call duration comprises:

identifying a second predicted call duration with the customer service agent for each of the one or more candidate time-periods;

identifying the subset of the one or more candidate time-periods having the second predicted call duration with the customer service agent within a duration percentage of the predicted call duration with the customer service agent; and

presenting the subset of the one or more candidate time-periods by the user computing entity of the caller.

5. The system of claim 4 , wherein identifying the second predicted call duration with the customer service agent for each of the one or more candidate time-periods comprises determining the second predicted call duration with the customer service agent based at least in part on call center dynamics during each of the one or more candidate time-periods.

6. The system of claim 1 , wherein accessing the calendar of the caller comprises:

receiving historical data associated with historical interactions between the caller and one or more customer service agents; and

executing at least one machine-learning model to identify the one or more candidate time-periods based at least in part on the historical data.

7. A computer-implemented method comprising:

receiving call-intent data for a caller during a voice-based interaction using a real-time communication channel with a user computing entity of the caller;

determining an unadjusted time estimate and a first confidence score associated with the unadjusted time estimate using the call-intent data as input to a first machine-learning time model;

in response to the first confidence score satisfying a first threshold, determining a time adjustment and a second confidence score associated with the time adjustment using an identity of a customer service agent assigned to the voice-based interaction of the caller as input to a second machine-learning time model;

in response to the second confidence score satisfying a second threshold, determining a predicted call duration with the customer service agent based at least in part on the unadjusted time estimate and the time adjustment; and

scheduling an interaction with the customer service agent to address a call-intent for the caller at least in part by:

accessing a calendar of the caller using a direct communication protocol of an executable application operating on the user computing entity of the caller, wherein the direct communication protocol enables data transfer from the user computing entity of the caller during the interaction;

identifying one or more candidate time-periods within the calendar of the caller having a time-period duration equal to or greater than the predicted call duration with the customer service agent, wherein the time-period duration begins when the interaction with the customer service agent is set to begin;

presenting at least a subset of the one or more candidate time-periods using an interactive user interface displayable on the user computing entity of the caller;

receiving input identifying a selected time-period of the one or more candidate time-periods; and

scheduling the interaction with the customer service agent to address the call-intent for the caller within the selected time-period.

8. The computer-implemented method of claim 7 , wherein receiving the call-intent data for the caller comprises:

receiving audio comprising a caller's voice for a call;

executing a voice-to-text conversion of the audio to generate transcription data; and

extracting one or more keywords from the transcription data.

9. The computer-implemented method of claim 7 , wherein accessing the calendar of the caller comprises one of:

receiving calendar data from the user computing entity of the caller, during the voice-based interaction; or

accessing a web-accessible calendar of the caller.

10. The computer-implemented method of claim 7 , wherein identifying the one or more candidate time-periods within the calendar of the caller for the interaction having the predicted call duration comprises:

identifying a second predicted call duration with the customer service agent for each of the one or more candidate time-periods;

identifying the subset of the one or more candidate time-periods having the second predicted call duration with the customer service agent within a duration percentage of the predicted call duration with the customer service agent; and

presenting the subset of the one or more candidate time-periods by the user computing entity of the caller.

11. The computer-implemented method of claim 10 , wherein identifying the second predicted call duration with the customer service agent for each of the one or more candidate time-periods comprises determining the second predicted call duration with the customer service agent based at least in part on call center dynamics during each of the one or more candidate time-periods.

12. The computer-implemented method of claim 7 , wherein accessing the calendar of the caller comprises:

receiving historical data associated with historical interactions between the caller and one or more customer service agents; and

executing at least one machine-learning model to identify the one or more candidate time-periods based at least in part on the historical data.

13. One or more non-transitory computer-readable storage media including instructions that, when executed by one or more processors, cause the one or more processors to:

receive call-intent data for a caller during a voice-based interaction using a real-time communication channel with a user computing entity of the caller;

determine an unadjusted time estimate and a first confidence score associated with the unadjusted time estimate using the call-intent data as input to a first machine-learning time model;

in response to the first confidence score satisfying a first threshold, determine a time adjustment and a second confidence score associated with the time adjustment using an identity of a customer service agent assigned to the voice-based interaction of the caller as input to a second machine-learning time model;

in response to the second confidence score satisfying a second threshold, determine a predicted call duration with the customer service agent based at least in part on the unadjusted time estimate and the time adjustment; and

schedule an interaction with the customer service agent to address a call-intent for the caller at least in part by:

accessing a calendar of the caller using a direct communication protocol of an executable application operating on the user computing entity of the caller, wherein the direct communication protocol enables data transfer from the user computing entity of the caller during the interaction;

identifying one or more candidate time-periods within the calendar of the caller having a time-period duration equal to or greater than the predicted call duration with the customer service agent, wherein the time-period duration begins when the interaction with the customer service agent is set to begin;

presenting at least a subset of the one or more candidate time-periods using an interactive user interface displayable on the user computing entity of the caller;

receiving input identifying a selected time-period of the one or more candidate time-periods; and

scheduling the interaction with the customer service agent to address the call-intent for the caller within the selected time-period.

14. The one or more non-transitory computer-readable storage media of claim 13 , wherein, to receive the call-intent data for the caller, the one or more processors are further caused to:

receive audio comprising a caller's voice for a call;

execute a voice-to-text conversion of the audio to generate transcription data; and

extract one or more keywords from the transcription data.

15. The one or more non-transitory computer-readable storage media of claim 13 , wherein accessing the calendar of the caller comprises one of:

receiving calendar data from the user computing entity of the caller, during the voice-based interaction; or

accessing a web-accessible calendar of the caller.

16. The one or more non-transitory computer-readable storage media of claim 13 , wherein identifying the one or more candidate time-periods within the calendar of the caller for the interaction having the predicted call duration comprises:

identifying a second predicted call duration with the customer service agent for each of the one or more candidate time-periods;

identifying the subset of the one or more candidate time-periods having the second predicted call duration with the customer service agent within a duration percentage of the predicted call duration with the customer service agent; and

presenting the subset of the one or more candidate time-periods by the user computing entity of the caller.

17. The one or more non-transitory computer-readable storage media of claim 16 , wherein identifying the second predicted call duration with the customer service agent for each of the one or more candidate time-periods comprises determining the second predicted call duration with the customer service agent based at least in part on call center dynamics during each of the one or more candidate time-periods.

18. The one or more non-transitory computer-readable storage media of claim 13 , wherein accessing the calendar of the caller comprises:

receiving historical data associated with historical interactions between the caller and one or more customer service agents; and

executing at least one machine-learning model to identify the one or more candidate time-periods based at least in part on the historical data.

Assignments (1)

Continuity (1)

References Cited (72)