Back to results
US 10,896,673 B1Granted
Authentication of impaired voices
AssigneeWells Fargo Bank, N.A.
InventorsAndrew J. Garner, IV, Tyua Larsen Fraser, Kimberly Ann MacInnis, Paul R. McMahon, Darrell Lee Suen, Zhong Wan
Life of this patent
Estimating the term…
Abstract
Systems and techniques for authentication of impaired voices are described herein. A voice profile may be generated for a user. An audio stream may be received including an authentication voice of the user. It may be determined that the authentication voice does not match a first set of authentication criteria. The audio stream may be compared to a second set of authentication criteria. The user may be authenticated based on the comparison.
Claims20
An authentication system for impaired voice authentication, the system comprising;
- at least one processor;
- and
- memory including instructions that, when executed by the at least one processor, cause the at least one processor to perform operations to:
- generate an initial voice profile for a user, the initial voice profile including acoustical properties of an initial voice of the user;
- collect a plurality of voice samples of the user over time;
- identify changes between the initial voice and the voice samples collected over time;
- generate an alternate voice profile for the user using the initial voice and the identified changes;
- receive an audio stream including an authentication voice of the user;
- determine that the authentication voice does not match a first set of authentication criteria, the first set of authentication criteria including features of the acoustical properties of the initial voice of the user and features of acoustical properties of the identified changes of the alternate voice profile;
- based on the determination, compare the audio stream to a second set of authentication criteria, wherein the second set of authentication criteria includes additional data elements derived from information included in the initial voice profile;
- determine an authentication level for the user, the authentication level based on a comparison between the authentication voice and the second set of authentication criteria, wherein the authentication level is determined based on a number of members of the second set of authentication criteria identified in the authentication voice;
- and
- authenticate the user for the authentication level based in part on the comparison between the authentication voice and the second set of authentication criteria and the comparison of the audio stream to the second set of authentication criteria, wherein the comparison of the audio stream to the second set of authentication criteria includes a comparison of the additional data elements derived from the information included in the initial voice profile to data elements of the audio stream.
10 more elements
At least one non-transitory machine readable medium including instructions for impaired voice authentication that, when executed by at least one processor, cause the at least one processor to perform operations to:
- generate initial voice profile for a user, the initial voice profile including acoustical properties of an initial voice of the user;
- collect a plurality of voice samples of the user over time;
- identify changes between the initial voice and the voice samples collected over time;
- generate an alternate voice profile for the user using the initial voice and the identified changes;
- receive an audio stream including an authentication voice of the user;
- determine that the authentication voice does not match a first set of authentication criteria, the first set of authentication criteria including features of the acoustical properties of the initial voice of the user and features of acoustical properties of the identified changes of the alternate voice profile;
- based on the determination, compare the audio stream to a second set of authentication criteria, wherein the second set of authentication criteria includes additional data elements derived from information included in the initial voice profile;
- determine an authentication level for the user, the authentication level based on a comparison between the authentication voice and the second set of authentication criteria, wherein the authentication level is determined based on a number of members of the second set of authentication criteria identified in the authentication voice;
- and
- authenticate the user for the authentication level based in part on the comparison between the authentication voice and the second set of authentication criteria and the comparison of the audio stream to the second set of authentication criteria, wherein the comparison of the audio stream to the second set of authentication criteria includes a comparison of the additional data elements derived from the information included in the initial voice profile to data elements of the audio stream.
7 more elements
A method for impaired voice authentication, the method comprising:
- generating an initial voice profile for a user, the initial voice profile including acoustical properties of an initial voice of the user;
- collecting a plurality of voice samples of the user over time;
- identifying changes between the initial voice and the voice samples collected over time;
- generating an alternate voice profile for the user using the initial voice and the identified changes;
- receiving, by a voice-enabled authentication system, an audio stream including an authentication voice of the user;
- determining that the authentication voice does not match a first set of authentication criteria of the voice-enabled authentication system, the first set of authentication criteria including features of the acoustical properties of the initial voice of the user and features of acoustical properties of the identified changes of the alternate voice profile;
- based on the determination, comparing the audio stream to a second set of authentication criteria of the voice-enabled authentication system, wherein the second set of authentication criteria includes additional data elements derived from information included in the initial voice profile;
- determining an authentication level for the user, the authentication level based on a comparison between the authentication voice and the second set of authentication criteria, wherein the authentication level is determined based on a number of members of the second set of authentication criteria identified in the authentication voice;
- and
- authenticating the user for the authentication level based in part on the comparison between the authentication voice and the second set of authentication criteria and the comparison of the audio stream to the second set of authentication criteria, wherein the comparison of the audio stream to the second set of authentication criteria includes a comparison of the additional data elements derived from the information included in the initial voice profile to data elements of the audio stream.
7 more elements
Dependencies are inferred from the claim text.