Multi-sample conversational voice verification

Patent metadata

Field	Value
Publication number	US-9443523-B2
Application number	US-201615012542-A
Country	US
Kind code	B2
Filing date	Feb 1, 2016
Priority date	Jun 15, 2012
Publication date	Sep 13, 2016
Grant date	Sep 13, 2016

Abstract

A system and method of verifying the identity of an authorized user in an authorized user group for enabling secure access to one or more services via a device includes receiving first voice information from a speaker through the device, calculating a confidence score based on a comparison of the first voice information with a stored voice model associated with the authorized user and specific to the authorized user, interpreting the first voice information as a specific service request, identifying a minimum confidence score for initiating the specific service request, determining whether or not the confidence score exceeds the minimum confidence score, and initiating the specific service request if the confidence score exceeds the minimum confidence score.

First claim

What is claimed is:

1. A computer-implemented method of verifying the identity of a speaker as an authorized user in an authorized user group to a level necessary for enabling secure access to one or more services via a device, the method comprising:
receiving first voice information comprising a specific service request from a speaker through the device before verifying the identity of the speaker as an authorized user;
calculating, by at least one processor, a confidence score that the speaker is an authorized user based on a comparison of the first voice information comprising the specific service request received before verifying the identity of the speaker as an authorized user with a stored voice model associated with the authorized user group and specific to the authorized user, wherein the at least one processor is included in at least one of the device and a server operatively coupled to the device through a communication network;
interpreting, by the at least one processor, the first voice information as a specific service request;
identifying, by the at least one processor, a type of service request comprising the specific service request;
identifying, by the at least one processor, a minimum confidence score for initiating the type of service request comprising the specific service request;
determining, by the at least one processor, whether or not the confidence score exceeds the minimum confidence score; and
if the confidence score exceeds the minimum confidence score, initiating the specific service request;
if the confidence score does not exceed the minimum confidence score, obtaining second voice information concerning the specific service request from the speaker through the device;
calculating a confidence score based on a comparison of a combination of the first voice information comprising the specific service request and the second voice information concerning the specific service request with the stored voice model associated with the authorized user group and specific to the authorized user; and repeating the determining step.

2. The computer-implemented method of claim 1 , further comprising constructing, by the at least one processor, the stored voice model based on voice information representing a set of service requests spoken by the speaker prior to receiving the first voice information.

3. The computer-implemented method of claim 2 , further comprising updating, by the at least one processor, the stored voice model based on the first voice information.

4. The computer-implemented method of claim 1 , further comprising storing, in the device, a plurality of minimum confidence scores each corresponding to a type of service request.

5. The computer-implemented method of claim 1 , further comprising prompting the speaker for the second voice information if the confidence score does not exceed the minimum confidence score.

6. The computer-implemented method of claim 5 , wherein prompting the speaker for the second voice information includes prompting the speaker to speak specific words based on the specific service request.

7. The computer-implemented method of claim 5 , wherein prompting the speaker for the second voice information includes:
looking up, in a database, options related to fulfillment of a type of service corresponding to the specific service request; and
constructing, by the at least one processor, the prompt based on at least one of the options.

8. The computer-implemented method of claim 1 , further comprising denying access if the confidence score does not exceed the minimum confidence score subsequent to receiving a predetermined amount of voice information from the speaker.

9. The computer-implemented method of claim 1 , further comprising interpreting, by the at least one processor, the second voice information as additional information related to the specific service request.

10. A system for verifying the identity of a speaker as an authorized user in an authorized user group to a level necessary for enabling secure access to one or more services via a device, the system comprising:
at least one processor included in at least one of the device and a server operatively coupled to the device through a communication network, the device configured to receive voice information from a speaker; and
a memory operatively coupled to the at least one processor, the memory including computer-executable instructions that when executed by the at least one processor cause the at least one processor to:
receive first voice information comprising a specific service request from a speaker through the device before verifying the identity of the speaker as an authorized user;
calculate a confidence score based on a comparison of the first voice information comprising the specific service request received before verifying the identity of the speaker as an authorized user with a stored voice model associated with the authorized user group and specific to the authorized user;
interpret the first voice information as a specific service request;
identify, by the at least one processor, a type of service request comprising the specific service request;
identify a minimum confidence score for initiating the type of service request comprising the specific service request;
determine whether or not the confidence score exceeds the minimum confidence score; and
if the confidence score exceeds the minimum confidence score, initiate the specific service request;
if the confidence score does not exceed the minimum confidence score, obtain second voice information concerning the specific service request from the speaker through the device; calculate a confidence score based on a comparison of a combination of the first voice information comprising the specific service request and the second voice information concerning the specific service request with the stored voice model associated with the authorized user group and specific to the authorized user; and repeat the determining step.

11. The system of claim 10 , wherein the memory further includes computer-executable instructions that when executed by the at least one processor cause the at least one processor to construct the stored voice model based on voice information representing a set of service requests spoken by the speaker prior to receiving the first voice information.

12. The system of claim 11 , wherein the memory further includes computer-executable instructions that when executed by the at least one processor cause the at least one processor to update the stored voice model based on the first voice information.

13. The system of claim 10 , wherein the memory further includes computer-executable instructions that when executed by the at least one processor cause the at least one processor to prompt the speaker for the second voice information if the confidence score does not exceed the minimum confidence score.

14. The system of claim 10 , wherein the memory further includes computer-executable instructions that when executed by the at least one processor cause the at least one processor to interpret the second voice information as additional information related to the specific service request.

15. A non-transitory computer-readable medium having stored thereon computer-executable instructions that when executed by at least one processor cause the at least one processor to:
receive first voice information comprising a specific service request from a speaker through a device before verifying the identity of the speaker as an authorized user;
calculate a confidence score that the speaker is an authorized user based on a com

Assignees

• Stanford Res Inst Int

Inventors

• Scheffer Nicolas
• Lei Yun
• Bercow Douglas A

Classifications

• G06F21/32

  using biometric data, e.g. fingerprints, iris scans or voiceprints · CPC title

• G10L17/22 Primary

  Interactive procedures; Man-machine interfaces · CPC title

• G06F2221/2113

  Multi-level security, e.g. mandatory access control · CPC title

• G06F16/24578

  using ranking · CPC title

• H04L63/0861

  using biometrical features, e.g. fingerprint, retina-scan (cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231) · CPC title