Verifying a user using speaker verification and a multimodal web-based interface

• US 20060190264A1
• Filed: 02/22/2005
• Published: 08/24/2006
• Est. Priority Date: 02/22/2005
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Jaramillo, David, McCobb, Gerald Matthew
• Granted Patent: US 8,725,514 B2
• Time in Patent Office: Days
• US Class Current: 704/270.1
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: H04M 2201/41 using speaker recognition
• CPC Class Codes: H04M 3/385 using speech signals

Abstract

A method of verifying a user identity using a Web-based multimodal interface can include sending, to a remote computing device, a multimodal markup language document that, when rendered by the remote computing device, queries a user for a user identifier and causes audio of the user'"'"'s voice to be sent to a multimodal, Web-based application. The user identifier and the audio can be received at about a same time from the client device. The audio can be compared with a voice print associated with the user identifier. The user at the remote computing device can be selectively granted access to the system according to a result obtained from the comparing step.

Claims

1. A method of verifying a user identity using a multimodal, Web-based interface comprising:

sending, to a remote computing device, a multimodal markup language document that, when rendered by the remote computing device, queries a user for a user identifier and causes audio of the user'"'"'s voice to sent to a multimodal, Web-based application;

receiving, at about a same time from the client device, the user identifier and the audio;

comparing the audio with a voice print associated with the user identifier; and

selectively granting the user at the remote computing device access to an online system according to a result obtained from said comparing step.

2. The method of claim 1, wherein the multimodal markup language document includes an instruction to stream the audio, said method further comprising, responsive to rendering the multimodal markup language document and interpreting the instruction to stream the audio, the remote computing device streaming the audio to the multimodal, Web-based application.

3. The method of claim 1, wherein the multimodal markup language document includes an instruction to record the audio, said method further comprising, responsive to rendering the multimodal markup language document and interpreting the record instruction, the remote computing device making a digital recording of the user'"'"'s voice.

4. The method of claim 1, wherein the multimodal markup language document specifies a script to be read aloud by the user.

5. The method of claim 4, wherein the script comprises, at least in part, randomly generated text.

6. The method of claim 4, wherein the audio is a digital representation of the user reading the script aloud.

7. The method of claim 6, said comparing step further comprising the multimodal Web-based application first sending, to a verification system, the audio, the script, and the user identifier.

8. The method of claim 1, further comprising:

first sending to the remote computing device a multimodal markup language document specifying an enrollment script;

receiving the user identifier and audio comprising a digital representation of the user speaking the enrollment script; and

enrolling the user into an online system by determining a voice print from the audio of the user speaking the enrollment script and associating the voice print with the user identifier.

9. A system for verifying a user identity comprising:

a verification system having a voice print database including a plurality of voice prints, each being associated with a user identifier;

a multimodal application configured to send a multimodal markup language document to a client device, wherein the multimodal markup language document specifies a random script and, upon execution, causes a user identifier and audio, comprising a digital representation of a speaker reading the random script aloud, to be sent back to the multimodal application; and

wherein the multimodal application submits the script, the audio, and the user identifier to said verification system for verifying the identity of the speaker by comparing the audio with a voice print selected from the plurality of voice prints that is associated with the user identifier.

10. The system of claim 9, wherein the audio and the user identifier are submitted to the multimodal application from the client device at approximately a same time.

11. The system of claim 9, wherein the multimodal markup language document includes an instruction to stream the audio, said method further comprising, responsive to rendering the multimodal markup language document and interpreting the instruction to stream the audio, the remote computing device streaming the audio to the multimodal application.

12. The system of claim 9, wherein the multimodal markup language document includes an instruction to record audio such that, responsive to rendering the multimodal markup language document and interpreting the record instruction, the client device records the speaker reading the script aloud.

13. A machine readable storage, having stored thereon a computer program having a plurality of code sections executable by a machine for causing the machine to perform the steps of:

sending, to a remote computing device, a multimodal markup language document that, when rendered by the remote computing device, queries a user for a user identifier and causes audio of the user'"'"'s voice to sent to a multimodal, Web-based application;

receiving, at about a same time from the client device, the user identifier and the audio;

comparing the audio with a voice print associated with the user identifier; and

selectively granting the user at the remote computing device access to an online system according to a result obtained from said comparing step.

14. The machine readable storage of claim 13, wherein the multimodal markup language document includes an instruction to stream the audio, said method further comprising, responsive to rendering the multimodal markup language document and interpreting the instruction to stream the audio, the remote computing device streaming the audio to the multimodal, Web-based application.

15. The machine readable storage of claim 13, wherein the multimodal markup language document includes an instruction to record audio, said method further comprising, responsive to rendering the multimodal markup language document and interpreting the record instruction, the remote computing device making a digital recording of the user'"'"'s voice.

16. The machine readable storage of claim 13, wherein the multimodal markup language document specifies a script to be read aloud by the user.

17. The machine readable storage of claim 16, wherein the script comprises, at least in part, randomly generated text.

18. The machine readable storage of claim 16, wherein the audio comprises a digital representation of the user reading the script aloud.

19. The machine readable storage of claim 18, said comparing step further comprising the Web-based multimodal application first sending, to a verification system, the audio, the script, and the user identifier.

20. The machine readable storage of claim 13, further comprising:

first sending to the remote computing device a multimodal markup language document specifying an enrollment script;

receiving the user identifier and audio comprising a digital representation of the user speaking the enrollment script; and

enrolling the user into an online system by determining a voice print from the audio of the user speaking the enrollment script and associating the voice print with the user identifier.