Granted Patent B2

US 12,431,135 · App. 18/310,929 · Granted Sep 30, 2025

Inventors: Saravanan Balasubramanian (Chennai, IN); Durga Prasad Kutthumolu (Hyderabad, IN); Venkata Karthik Ryali (Hyderabad, IN)

Assignee: Bank of America Corporation

G10L15/22G06F9/453G06F9/454G06F16/243G06F40/58G10L15/005G10L15/063G10L15/30G10L15/32H04L63/08G10L2015/0635

Abstract

Claims (150)

1. A system comprising:

a metaverse server comprising:

a first memory configured to store:

user credential of a plurality of users; and

a first processor communicatively coupled to the first memory, wherein the first processor is configured to:

receive first user credentials of a first user from a first user device;

authenticate the first user based on the first user credentials;

generate a metaverse session;

generate a first avatar of the first user;

generate a virtual advisor avatar;

allow the first user access to the metaverse session using the first avatar;

receive a voice query of the first user from the first user device;

send the voice query to a decentralized computing system;

in response to sending the voice query to the decentralized computing system, receive a first voice response from the decentralized computing system; and

communicate the first voice response to the first user using the virtual advisor avatar; and

the decentralized computing system communicatively coupled to the metaverse server, the decentralized computing system comprising a plurality of second processors, wherein the plurality of second processors are configured to implement a plurality of bots, the plurality of bots comprising a primary bot and a plurality of secondary bots, wherein the plurality of second processors of the decentralized computing system are configured to:

receive, by the primary bot, the voice query from the metaverse server;

analyze, by the primary bot, the voice query to determine:

a preferred language/dialect of the first user;

an issue and a domain of the voice query; and

a respective probability score for each secondary bot;

identify a secondary bot with a highest probability score, the secondary bot with the highest probability score being configured to respond to the voice query in the preferred language/dialect of the first user;

search a language/knowledge server for a response corresponding to the issue and the domain;

in response to determining that the language/knowledge server includes the response corresponding to the issue and the domain, retrieve the response from the language/knowledge server;

determine a language/dialect of the response;

in response to determining that the language/dialect of the response matches the preferred language/dialect of the first user, generate, by the secondary bot with the highest probability score, the first voice response from the retrieved response, wherein the first voice response is in the preferred language/dialect of the first user; and

send the first voice response to the metaverse server.

2. The system of claim 1 , wherein the plurality of second processors of the decentralized computing system is further configured to:

in response to determining that the language/dialect of the response does not match the preferred language/dialect of the first user:

translate, by the primary bot, the response into the preferred language/dialect of the first user; and

generate, by the secondary bot with the highest probability score, the first voice response from the translated response.

3. The system of claim 1 , wherein the first processor of the metaverse server is further configured to:

in response to determining that the language/knowledge server does not include the response corresponding to the issue and the domain:

determine a location of the first user;

determine a regional center based on the location;

determine an available advisor within the regional center;

notify the available advisor that the first user requires assistance;

allow the available advisor access to the metaverse session using the virtual advisor avatar;

in response to determining that the preferred language/dialect of the available advisor is same as the preferred language/dialect of the first user:

communicate the voice query to the available advisor;

receive a second voice response from the available advisor; and

communicate the second voice response to the first user using the virtual advisor avatar.

4. The system of claim 3 , wherein the first processor of the metaverse server is further configured to:

update the language/knowledge server based on the second voice response.

5. The system of claim 3 , wherein:

the plurality of second processors of the decentralized computing system are further configured to:

translate, by the primary bot, the voice query into the preferred language/dialect of the available advisor; and

translate, by the primary bot, a third voice response into the preferred language/dialect of the first user; and

the first processor of the metaverse server is further configured to:

in response to determining that the preferred language/dialect of the available advisor is not same as the preferred language/dialect of the first user:

receive the translated voice query from the decentralized computing system;

communicate the translated voice query to the available advisor;

receive the third voice response from the available advisor;

send the third voice response to the decentralized computing system;

receive the translated third voice response from the decentralized computing system; and

communicate the translated third voice response to the first user using the virtual advisor avatar.

6. The system of claim 5 , wherein the first processor of the metaverse server is further configured to:

update the language/knowledge server based on the translated third voice response.

7. The system of claim 1 , wherein the first user accesses the metaverse session from a metaverse kiosk.

8. A method comprising:

receiving first user credentials of a first user from a first user device;

authenticating the first user based on the first user credentials;

generating a metaverse session;

generating a first avatar of the first user;

generating a virtual advisor avatar;

allowing the first user access to the metaverse session using the first avatar;

receiving a voice query of the first user from the first user device;

sending the voice query to a decentralized computing system, the decentralized computing system comprising a plurality of second processors, wherein the plurality of second processors are configured to implement a plurality of bots, the plurality of bots comprising a primary bot and a plurality of secondary bots;

analyzing, by the primary bot, the voice query to determine:

a preferred language/dialect of the first user;

an issue and a domain of the voice query; and

a respective probability score for each secondary bot;

identifying a secondary bot with a highest probability score, the secondary bot with the highest probability score being configured to respond to the voice query in the preferred language/dialect of the first user;

searching a language/knowledge server for a response corresponding to the issue and the domain;

in response to determining that the language/knowledge server includes the response corresponding to the issue and the domain, retrieving the response from the language/knowledge server;

determining a language/dialect of the response;

in response to determining that the language/dialect of the response matches the preferred language/dialect of the first user, generating, by the secondary bot with the highest probability score, a first voice response from the retrieved response, wherein the first voice response is in the preferred language/dialect of the first user; and

communicating the first voice response to the first user using the virtual advisor avatar.

9. The method of claim 8 , further comprising:

in response to determining that the language/dialect of the response does not match the preferred language/dialect of the first user:

translating, by the primary bot, the response into the preferred language/dialect of the first user; and

generating, by the secondary bot with the highest probability score, the first voice response from the translated response.

10. The method of claim 8 , further comprising:

in response to determining that the language/knowledge server does not include the response corresponding to the issue and the domain:

determining a location of the first user;

determining a regional center based on the location;

determining an available advisor within the regional center;

notifying the available advisor that the first user requires assistance;

allowing the available advisor access to the metaverse session using the virtual advisor avatar; and

in response to determining that the preferred language/dialect of the available advisor is same as the preferred language/dialect of the first user:

communicating the voice query to the available advisor;

receiving a second voice response from the available advisor; and

communicating the second voice response to the first user using the virtual advisor avatar.

11. The method of claim 10 , further comprising:

updating the language/knowledge server based on the second voice response.

12. The method of claim 10 , further comprising:

in response to determining that the preferred language/dialect of the available advisor is not same as the preferred language/dialect of the first user:

translating, by the primary bot, the voice query into the preferred language/dialect of the available advisor;

communicating the translated voice query to the available advisor;

receiving a third voice response from the available advisor;

translating, by the primary bot, the third voice response into the preferred language/dialect of the first user; and

communicating the translated third voice response to the first user using the virtual advisor avatar.

13. The method of claim 12 , further comprising:

updating the language/knowledge server based on the translated third voice response.

14. The method of claim 8 , wherein the first user accesses the metaverse session from a metaverse kiosk.

15. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to:

receive first user credentials of a first user from a first user device;

authenticate the first user based on the first user credentials;

generate a metaverse session;

generate a first avatar of the first user;

generate a virtual advisor avatar;

allow the first user access to the metaverse session using the first avatar;

receive a voice query of the first user from the first user device;

send the voice query to a decentralized computing system, the decentralized computing system comprising a plurality of second processors, wherein the plurality of second processors are configured to implement a plurality of bots, the plurality of bots comprising a primary bot and a plurality of secondary bots;

analyze, by the primary bot, the voice query to determine:

a preferred language/dialect of the first user;

an issue and a domain of the voice query; and

a respective probability score for each secondary bot;

identify a secondary bot with a highest probability score, the secondary bot with the highest probability score being configured to respond to the voice query in the preferred language/dialect of the first user;

search a language/knowledge server for a response corresponding to the issue and the domain;

in response to determining that the language/knowledge server includes the response corresponding to the issue and the domain, retrieve the response from the language/knowledge server;

determine a language/dialect of the response;

in response to determining that the language/dialect of the response matches the preferred language/dialect of the first user, generate, by the secondary bot with the highest probability score, a first voice response from the retrieved response, wherein the first voice response is in the preferred language/dialect of the first user; and

communicate the first voice response to the first user using the virtual advisor avatar.

16. The non-transitory computer-readable medium of claim 15 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:

in response to determining that the language/dialect of the response does not match the preferred language/dialect of the first user:

translate, by the primary bot, the response into the preferred language/dialect of the first user; and

generate, by the secondary bot with the highest probability score, the first voice response from the translated response.

17. The non-transitory computer-readable medium of claim 15 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:

in response to determining that the language/knowledge server does not include the response corresponding to the issue and the domain:

determine a location of the first user;

determine a regional center based on the location;

determine an available advisor within the regional center;

notify the available advisor that the first user requires assistance;

allow the available advisor access to the metaverse session using the virtual advisor avatar; and

in response to determining that the preferred language/dialect of the available advisor is same as the preferred language/dialect of the first user:

communicate the voice query to the available advisor;

receive a second voice response from the available advisor; and

communicate the second voice response to the first user using the virtual advisor avatar.

18. The non-transitory computer-readable medium of claim 17 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:

update the language/knowledge server based on the second voice response.

19. The non-transitory computer-readable medium of claim 17 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:

in response to determining that the preferred language/dialect of the available advisor is not same as the preferred language/dialect of the first user:

translate, by the primary bot, the voice query into the preferred language/dialect of the available advisor;

communicate the translated voice query to the available advisor;

receive a third voice response from the available advisor;

translate, by the primary bot, the third voice response into the preferred language/dialect of the first user; and

communicate the translated third voice response to the first user using the virtual advisor avatar.

20. The non-transitory computer-readable medium of claim 19 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:

update the language/knowledge server based on the translated third voice response.

Assignments (1)

Continuity (1)

References Cited (22)