Systems and methods for cluster-based voice verification

• US 10,574,812 B2
• Filed: 08/01/2019
• Issued: 02/25/2020
• Est. Priority Date: 02/08/2018
• Status: Active Grant

• Current Assignee: Capital One Services LLC (Capital One Financial Corporation)
• Original Assignee: Capital One Services LLC (Capital One Financial Corporation)
• Inventors: Walters, Austin, Goodsitt, Jeremy, Abdi Taghi Abad, Fardin
• Primary Examiner(s): Milord, Marceau
• Application Number:
  US16/529,203
• Publication Number:
  US 20190356776A1
• Time in Patent Office: 208 Days
• Field of Search: 455415, 4554141, 4554144
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
• CPC Class Codes: G10L 25/06 the extracted parameters being correlation coefficients
• CPC Class Codes: G10L 25/27 characterised by the analysis technique
• CPC Class Codes: G10L 25/30 using neural networks
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: H04M 2201/41 using speaker recognition speaker recognition per se G10L17/00
• CPC Class Codes: H04M 2203/556 Statistical analysis and interpretation
• CPC Class Codes: H04M 2203/6045 Identity confirmation
• CPC Class Codes: H04M 3/42042 Notifying the called party of information on the calling party details within substation equipment H04M1/57, signalling details H04Q3/72

Abstract

Systems for caller identification and authentication may include an authentication server. The authentication server may be configured to receive audio data including speech of a plurality of telephone calls, use audio data for at least a subset of the plurality of telephone calls to store a plurality of known characteristics each associated with a specific demographic, and/or use audio data for at least one of the plurality of telephone calls to identify the telephone caller making the telephone call based on determining a most similar known characteristic of the plurality of known characteristics to the audio data of the caller.

Claims

1. A method comprising:

receiving, by, a processor of an authentication server, audio data including speech of a user;

analyzing, by the processor, the audio data to identify at least one characteristic of the speech of the user;

associating, by the processor, the at least one characteristic to a cluster based on a comparison with a plurality of known characteristics, each known characteristic being associated with at least one cluster;

receiving, by the processor, data indicative of a purported demographic of the user;

comparing, by the processor, the data indicative of the purported demographic to data indicative of the cluster; and

identifying, by the processor, the user as;

likely belonging to the purported demographic in response to determining the data indicative of the purported demographic matches the data indicative of the at least one cluster, and unlikely to belong to the purported demographic in response to determining the data indicative of the purported demographic matches data indicative of a different cluster.

2. The method of claim 1, wherein:

the at least one characteristic of the speech of the user comprises a plurality of words;

each known characteristic comprises a plurality of associated words; and

the associating comprises determining a similarity of the plurality of words and the plurality of associated words of the associated cluster.

3. The method of claim 2, wherein:

the at least one characteristic of the speech of the user comprises an occurrence frequency for each of the plurality of words;

each known characteristic comprises an occurrence frequency for each of the plurality of associated words; and

the associating comprises determining a similarity of the occurrence frequency for each of the plurality of words and the occurrence frequency for each of the plurality of associated words of the associated cluster.

4. The method of claim 2, wherein:

the at least one characteristic of the speech of the user further comprises at least one acoustic characteristic;

each known characteristic further comprises at least one acoustic characteristic; and

the analyzing of the audio data to identify at least one acoustic characteristic of the speech of the user comprises;

correlating each of a plurality of portions of an acoustic or frequency component of the audio data with each of at least a subset of the plurality of words; and

determining the at least one acoustic characteristic for how the user says at least one of the subset of the plurality of words based on the portion of the acoustic or frequency component of the audio data correlated with the at least one of the subset of the plurality of words.

5. The method of claim 1, wherein:

the at least one characteristic of the speech of the user comprises at least one acoustic characteristic;

each known characteristic comprises at least one acoustic characteristic; and

the associating comprises determining a similarity of the at least one acoustic characteristic and the at least one acoustic characteristic of the associated cluster.

6. The method of claim 1, wherein:

the data indicative of the at least one of the plurality of users comprises current individual data and historical individual data;

determining the data indicative of the purported demographic matches the data indicative of the at least one of the plurality of users comprises determining at least one of the current individual data and the historical individual data matches the at least one of the plurality of users associated with the associated cluster; and

determining the data indicative of the purported demographic matches data indicative of a different at least one of the plurality of users comprises determining at least one of the current individual data and the historical individual data matches the at least one user associated with the known characteristic different from the associated cluster.

7. The method of claim 1, further comprising:

receiving, by the processor, a threat score for the user;

wherein the identifying, by the processor, the user as likely belonging to the purported demographic comprises lowering the threat score or maintaining the threat score as received.

8. The method of claim 1, further comprising:

receiving, by the processor, a threat score for the user;

wherein the identifying, by the processor, the user as unlikely to belong to the purported demographic comprises raising the threat score.

9. A system for user authentication, the system comprising:

a recorder configured to record audio data of speech spoken by a user;

an authentication server comprising a processor and a non-transitory memory, the memory storing instructions that, when executed by the processor, cause the processor to perform processing comprising;

receiving audio data including speech of a plurality of users;

using audio data for at least a subset of the plurality of users to store a plurality of known characteristics, each known characteristic being associated with at least one cluster, the storing comprising;

for each of the subsets of the plurality of users, determining identifying data for each user, and analyzing the audio data to identify at least one characteristic of the speech of the user, and

storing the at least one characteristic of the speech of each user included in the plurality of users based on the identifying data for the user as the known characteristic; and

using audio data for at least one of the plurality of users to identify the user, the identifying comprising;

analyzing the audio data to identify at least one characteristic of the speech of the user,

associating the at least one characteristic to a cluster based on a comparison with a plurality of known characteristics, each known characteristic being associated with at least one cluster,

receiving data indicative of a purported demographic of the user, and

identifying the user as;

likely belonging to the purported demographic in response to determining the data indicative of the purported demographic matches the data indicative of the at least one of the plurality of users, or

unlikely to belong to the purported demographic in response to determining the data indicative of the purported demographic matches data indicative of a different at least one of the plurality of users.

10. The system of claim 9, wherein:

the at least one characteristic of the speech of the user comprises a plurality of words;

each known characteristic comprises a plurality of associated words; and

the associating comprises determining a similarity of the plurality of words and the plurality of associated words of the associated cluster.

11. The system of claim 10, wherein:

the at least one characteristic of the speech of the user comprises an occurrence frequency for each of the plurality of words;

each known characteristic comprises an occurrence frequency for each of the plurality of associated words; and

the associating comprises determining a similarity of the occurrence frequency for each of the plurality of words and the occurrence frequency for each of the plurality of associated words of the associated cluster.

12. The system of claim 10, wherein:

the at least one characteristic of the speech of the user further comprises at least one acoustic characteristic;

each known characteristic further comprises at least one acoustic characteristic; and

the analyzing of the audio data to identify at least one acoustic characteristic of the speech of the user comprises;

correlating each of a plurality of portions of an acoustic or frequency component of the audio data with each of at least a subset of the known characteristics; and

determining the at least one acoustic characteristic for how the user says at least one of the subset of the plurality of words based on the portion of the acoustic or frequency component of the audio data correlated with the at least one of the subsets of the known characteristics.

13. The system of claim 9, wherein:

the at least one characteristic of the speech of the user comprises at least one acoustic characteristic;

each known characteristic comprises at least one acoustic characteristic; and

the associating comprises determining a similarity of the at least one acoustic characteristic and the at least one acoustic characteristic of the associated cluster.

14. The system of claim 9, wherein:

the processing further comprises receiving a threat score for the user; and

the identifying the user as unlikely to have the purported demographic comprises affecting the threat score.

15. A non-transitory computer readable medium storing instructions that, when executed by a processor, cause the processor to perform processing comprising:

receiving audio data including speech of a user;

analyzing the audio data to identify at least one characteristic of the speech of the user;

associating the at least one characteristic to a cluster based on a comparison with a plurality of known characteristics, each known characteristic being associated with at least one cluster;

receiving data indicative of a purported demographic of the user;

comparing the data indicative of the purported demographic to data indicative of the cluster; and

identifying the user as;

likely belonging to the purported demographic in response to determining the data indicative of the purported demographic matches the data indicative of the at least one cluster, and

unlikely to belong to the purported demographic in response to determining the data indicative of the purported demographic matches data indicative of a different cluster.

16. The computer readable medium of claim 15, wherein:

the at least one characteristic of the speech of the user comprises a plurality of words;

each known characteristic comprises a plurality of associated words; and

the associating comprises determining a similarity of the plurality of words and the plurality of associated words of the associated cluster.

17. The computer readable medium of claim 16, wherein:

the at least one characteristic of the speech of the user comprises an occurrence frequency for each of the plurality of words;

each known characteristic comprises an occurrence frequency for each of the plurality of associated words; and

the associating comprises determining a similarity of the occurrence frequency for each of the plurality of words and the occurrence frequency for each of the plurality of associated words of the associated cluster.

18. The computer readable medium of claim 16, wherein:

the at least one characteristic of the speech of the user further comprises at least one acoustic characteristic;

each known characteristic further comprises at least one acoustic characteristic; and

the analyzing of the audio data to identify at least one acoustic characteristic of the speech of the user comprises;

correlating each of a plurality of portions of an acoustic or frequency component of the audio data with each of at least a subset of the plurality of words; and

determining the at least one acoustic characteristic for how the user says at least one of the subset of the plurality of words based on the portion of the acoustic or frequency component of the audio data correlated with the at least one of the subset of the plurality of words.

19. The computer readable medium of claim 15, wherein:

the at least one characteristic of the speech of the user comprises at least one acoustic characteristic;

each known characteristic comprises at least one acoustic characteristic; and

the associating comprises determining a similarity of the at least one acoustic characteristic and the at least one acoustic characteristic of the associated cluster.

20. The computer readable medium of claim 15, wherein:

the data indicative of the at least one of the plurality of users comprises current individual data and historical individual data;

determining the data indicative of the purported demographic matches the data indicative of the at least one of the plurality of users comprises determining at least one of the current individual data and the historical individual data matches the at least one of the plurality of users associated with the associated cluster; and

determining the data indicative of the purported demographic matches data indicative of a different at least one of the plurality of users comprises determining at least one of the current individual data and the historical individual data matches the at least one users associated with the known characteristic different from the associated cluster.