Granted Patent B2

US 12,413,664 · App. 18/223,815 · Granted Sep 9, 2025

Inventors: Manu K M (Bangalore, IN); Balaji Sankar Kumar (Bangalore, IN); Vidya Chandrashekar (Bangalore, IN); vamshi dondapati (Hyderabad, IN); Arun Aravind (Mahe, IN); Akshat Dixit (Lucknow, IN); Arun Sabaresh Anantha Narayanan (Bangalore, IN)

Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION

H04M3/2281G10L15/1815H04L9/088H04L9/30H04M3/2218H04M2201/40

Abstract

Claims (60)

1. A computer-implemented method comprising:

converting to text data in real-time, by a voice-to-text converter, voice data of a conversation occurring in real-time between a first party and a second party;

extracting, by an extractor, an intent and an entity from the text data, wherein the intent comprises a request by the first party to the second party, and wherein the entity comprises a category of information being requested;

predicting, by a predictor, based on the request and the category of information being requested, a sensitive information that is likely to be disclosed in the real-time conversation;

comparing, by an intersector, the text data in real-time to the predicted sensitive information;

determining, by the intersector, responsive to the comparing, whether the text data includes the predicted sensitive information; and

intersecting, by the intersector, responsive to a determination that the text data includes the predicted sensitive information, the voice data.

2. The method of claim 1 , wherein the voice data is associated with a speaker of a telephonic conversation, further comprising:

intersecting the voice data includes preventing a transmission of the voice data to a listener in the telephonic conversation.

3. The method of claim 1 , further comprising:

retrieving a key corresponding to an identifier associated with the sensitive information, and a value corresponding to the sensitive information; and

predicting the sensitive information by determining whether the text data matches at least one of the key or the value.

4. The method of claim 3 , further comprising:

decomposing the predicted sensitive information into a key-value pair; and

storing the key-value pair.

5. The method of claim 1 , further comprising generating a sensitive information warning.

6. The method of claim 1 , further comprising generating a speaker agreement.

7. The method of claim 1 , further comprising:

determining a risk score associated with a telephonic conversation; and

generating a risk warning, responsive to a determination that the risk score meets a predetermined threshold.

8. The method of claim 7 , wherein the risk score is based on at least one of a call history, a call frequency, and a call time.

9. The method of claim 1 , further comprising:

deactivating, responsive to a user selection, at least one of the voice-to-text converter, the extractor, the predictor, and the intersector.

10. The method of claim 1 , further comprising:

comparing the entity to a key in a first key-value pair, wherein the entity is extracted from a first portion of voice data;

anticipating, responsive to the entity matching the key, that a potential voice response to the first portion of the voice data has a likelihood of including a first sensitive information corresponding to the first key-value pair; and

causing, responsive to the anticipating, a manipulation of a communication channel carrying the voice data.

11. The method of claim 1 , further comprising:

comparing a first portion of the text data with a first key-value pair;

anticipating, responsive to the first portion of the text data at least partially matching a portion of the first key-value pair, that an upcoming portion of the voice data has a likelihood of including a first sensitive information corresponding to the first key-value pair; and

causing a response to the anticipation, a manipulation of a communication channel carrying voice data.

12. A computer program product comprising one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable by a processor to cause the processor to perform operations comprising:

converting to text data in real-time, by a voice-to-text converter, voice data of a conversation occurring in real-time between a first party and a second party;

extracting, by an extractor, an intent and an entity from the text data, wherein the intent comprises a request by the first party to the second party, and wherein the entity comprises a category of information being requested;

predicting, by a predictor, based on the request and the category of information being requested, a sensitive information that is likely to be disclosed in the real-time conversation;

comparing, by an intersector, the text data in real-time to the predicted sensitive information;

determining, by the intersector, responsive to the comparing, whether the text data includes the predicted sensitive information; and

intersecting, by the intersector, responsive to a determination that the text data includes the predicted sensitive information, the voice data.

13. The computer program product of claim 12 , wherein the voice data is associated with a speaker of a telephonic conversation, further comprising:

intersecting the voice data includes preventing a transmission of the voice data to a listener in the telephonic conversation.

14. The computer program product of claim 12 , further comprising:

retrieving a key corresponding to an identifier associated with the sensitive information, and a value corresponding to the sensitive information; and

predicting the sensitive information by determining whether the text data matches at least one of the key or the value.

15. The computer program product of claim 14 , further comprising:

decomposing the predicted sensitive information into a key-value pair; and

storing the key-value pair.

16. The computer program product of claim 12 , further comprising generating a sensitive information warning.

17. The computer program product of claim 12 , further comprising generating a speaker agreement.

18. A computer system comprising a processor and one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable by the processor to cause the processor to perform operations comprising:

converting to text data in real-time, by a voice-to-text converter, voice data of a conversation occurring in real-time between a first party and a second party;

extracting, by an extractor, an intent and an entity from the text data, wherein the intent comprises a request by the first party to the second party, and wherein the entity comprises a category of information being requested;

predicting, by a predictor, based on the request and the category of information being requested, a sensitive information that is likely to be disclosed in the real-time conversation;

comparing, by an intersector, the text data in real-time to the predicted sensitive information;

determining, by the intersector, responsive to the comparing, whether the text data includes the predicted sensitive information; and

intersecting, by the intersector, responsive to a determination that the text data includes the predicted sensitive information, the voice data.

19. The computer system of claim 18 , wherein the voice data is associated with a speaker of a telephonic conversation, further comprising:

intersecting the voice data includes preventing a transmission of the voice data to a listener in the telephonic conversation.

20. The computer system of claim 18 , further comprising:

retrieving a key corresponding to an identifier associated with the sensitive information, and a value corresponding to the sensitive information; and

predicting the sensitive information by determining whether the text data matches at least one of the key or the value.

Assignments (1)

Continuity (1)

References Cited (25)

Cited By (1)

US 12,614,034