Granted Patent B1

US 12,412,568 · App. 18/321,916 · Granted Sep 9, 2025

Inventors: Wei Dai (Hangzhou, CN); Maikl Adly Abdel-Malek Eskander (Seattle, WA); Fengcai Sun (Hangzhou, CN)

Assignee: Zoom Communications, Inc.

G10L15/18G10L15/22G10L15/30H04M3/5183G06Q10/1095H04M2201/40

Abstract

Claims (62)

1. A method, comprising:

accessing, during an engagement with a contact center performed by a software agent configured to generate and output automated responses to prompts provided over one or more communication modalities, a prompt provided by a contact center user device;

determining, using a natural language processing engine of the software agent, to use an application programming interface call based on the prompt, the natural language processing engine comprising a generative pre-trained transformer trained to generate natural language responses to natural language prompts;

transmitting an access request representing the application programming interface call to a governance engine configured to generate data structures responsive to application programming interface calls in a format readable by the software agent;

receiving a data structure responsive to the access request from the governance engine;

generating, using the natural language processing engine, a natural language response based on the data structure; and

transmitting the natural language response to the contact center user device.

2. The method of claim 1 , wherein determining to use the application programming interface call based on the prompt comprises:

transmitting, to the contact center user device, a request for additional information to generate the application programming interface call;

receiving, from the contact center user device, natural language data in response to the request; and

generating the access request using the natural language data.

3. The method of claim 1 , wherein determining to use the application programming interface call based on the prompt comprises:

transmitting, to the contact center user device, a request for additional information to generate the application programming interface call;

receiving, from the contact center user device, natural language data in response to the request;

identifying a second prompt based on the natural language data; and

generating, by the software agent, a response to the second prompt in parallel with generating the natural language response to the prompt.

4. The method of claim 1 , wherein determining to use the application programming interface call based on the prompt comprises:

transmitting, to the contact center user device, a request for additional information to generate the application programming interface call;

receiving, from the contact center user device, natural language data in response to the request;

identifying a second prompt based on the natural language data; and

generating, by the software agent, a response to the second prompt after transmitting the natural language response to the contact center user device.

5. The method of claim 1 , wherein the governance engine generates the application programming interface call and receives a response to the application programming interface call, wherein the data structure represents the response to the application programming interface call in a format that is readable by the software agent.

6. The method of claim 1 , wherein the application programming interface call comprises a retrieval of data from a data repository, wherein the data comprises at least one of user information, a user account status, a user transaction history, product information, or real-time data.

7. The method of claim 1 , wherein the application programming interface call comprises processing a payment or verifying a user identity.

8. The method of claim 1 , wherein the application programming interface call comprises creating a support ticket.

9. The method of claim 1 , wherein the application programming interface call comprises scheduling an appointment.

10. The method of claim 1 , wherein the generative pre-trained transformer is trained, at least in part, using data of an entity using the contact center engagement.

11. A non-transitory computer-readable medium storing instructions operable to cause one or more processors to perform operations comprising:

accessing, during an engagement with a contact center performed by a software agent configured to generate and output automated responses to prompts provided during the engagement over one or more communication modalities, a prompt provided by a contact center user device;

determining, using a natural language processing engine of the software agent, to use an application programming interface call based on the prompt, the natural language processing engine comprising a generative pre-trained transformer trained to generate natural language responses to natural language prompts;

transmitting an access request representing the application programming interface call to a governance engine configured to generate data structures responsive to application programming interface calls in a format readable by the software agent;

receiving a data structure responsive to the access request from the governance engine;

generating, using the natural language processing engine, a natural language response based on the data structure; and

transmitting the natural language response to the contact center user device.

12. The non-transitory computer-readable medium of claim 11 , wherein determining to use the application programming interface call based on the prompt comprises:

transmitting, to the contact center user device, a request for information to generate the application programming interface call;

receiving, from the contact center user device, natural language text or natural language speech in response to the request; and

generating the access request using the natural language text or the natural language speech.

13. The non-transitory computer-readable medium of claim 11 , wherein determining to use the application programming interface call based on the prompt comprises:

transmitting, to the contact center user device, a request for information to generate the application programming interface call;

receiving, from the contact center user device, natural language text in response to the request;

identifying a second prompt based on the natural language text; and

generating, by the software agent, a response to the second prompt in parallel with generating the natural language response to the prompt.

14. The non-transitory computer-readable medium of claim 11 , wherein determining to use the application programming interface call based on the prompt comprises:

transmitting, to the contact center user device, a request for information to generate the application programming interface call;

receiving, from the contact center user device, natural language text in response to the request;

identifying a second prompt based on the natural language text; and

generating, by the software agent, a response to the second prompt after transmitting the natural language response to the contact center user device.

15. The non-transitory computer-readable medium of claim 11 , wherein the governance engine generates the application programming interface call and receives a response to the application programming interface call, wherein the data structure represents the response to the application programming interface call.

16. The non-transitory computer-readable medium of claim 11 , wherein the application programming interface call comprises a retrieval of data from a data repository, wherein the data comprises at least one of a user account status, a user transaction history, product information, or real-time data.

17. The non-transitory computer-readable medium of claim 11 , wherein the application programming interface call comprises processing a payment.

18. The non-transitory computer-readable medium of claim 11 , wherein the application programming interface call comprises booking an appointment.

19. The non-transitory computer-readable medium of claim 11 , wherein the generative pre-trained transformer is trained, at least in part, using data of an entity using the contact center to communicate with end-users of products or services of the entity.

20. An apparatus comprising:

a memory; and

a processor configured to execute instructions stored in the memory to:

access, during an engagement with a contact center performed by a software agent configured to generate and output automated responses to prompts provided during the engagement over one or more communication modalities, a prompt provided by a contact center user device;

determine, using a natural language processing engine of the software agent, to use an application programming interface call based on the prompt, the natural language processing engine comprising a generative pre-trained transformer trained to generate natural language responses to natural language prompts;

transmit an access request representing the application programming interface call to a governance engine configured to generate data structures responsive to application programming interface calls in a format readable by the software agent;

receive a data structure responsive to the access request from the governance engine;

generate, using the natural language processing engine, a natural language response based on the data structure; and

transmit the natural language response to the contact center user device.

Assignments (2)

References Cited (18)