VERIFYING A USER

• US 20130185071A1
• Filed: 07/17/2012
• Published: 07/18/2013
• Est. Priority Date: 12/23/2011
• Status: Active Grant

• Current Assignee: National ICT Australia Limited
• Original Assignee: National ICT Australia Limited
• Inventors: Chen, Fang, Yin, Bo, MacDonald, Kim
• Granted Patent: US 10,008,206 B2
• Time in Patent Office: Days
• US Class Current: 704/244
• CPC Class Codes: G06F 18/295 Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities
• CPC Class Codes: G10L 17/04 Training, enrolment or model building

Abstract

Verifying a user, such as, but not limited to, a user who answered questions for an unproctered test for employment. A representation of a transition pattern is stored (200). Time series voice data is received from a user responding to a sequence of questions (72) having multiple levels of difficulty. The transition pattern in the voice data (200) is based on a plurality of comparisons (503) of subsets (500) of the voice data corresponding to responses to questions having one or more different levels of difficulty. A further transition pattern (200) is determined from voice data in the same way but based on a shorter sequence of questions (78). The user is verified based on a comparison (78) of this further transition pattern to a previously stored transition pattern.

Claims

1. A computer implemented method to determine verifying data of a user, the system comprising:

receiving time series voice data from the user responding to a sequence of questions having multiple levels of difficulty;

determining a transition pattern in the voice data based on a plurality of comparisons of subsets of the voice data corresponding to responses to questions having one or more different levels of difficulty; and

storing a representation of the transition pattern for use in future verification of user.

2. The computer implemented method according to claim 1, wherein the sequence of questions form an unproctored test of the user for suitability or qualification to a role that the user is being tested for.

3. The computer implemented method according to claim 1, wherein for each comparison of subsets of voice data, at least one subset corresponds to a response to a question having a level of difficulty significantly different to the other subsets.

4. The computer implemented method according to claim 1, wherein the method of determining the transition pattern comprises determining from each subset one or more of the following features:

pitch,

first formant,

second formant,

third formant,

glottal excitation, and

place of articulation.

5. The computer implemented method according to claim 1, wherein the comparison of subsets is based on a comparison of one or more features of the subsets, and the method comprises determining a probability distribution function of one or more features of each subset.

6. The computer implemented method according to claim 5, wherein the comparison of subsets is based on a distance measure of the respective probability distribution function.

7. The computer implemented method according to claim 6, wherein the questions are in substantially increasing or decreasing order of difficulty such that the distance measures are combined to form a time series vector.

8. The computer implemented method according to claim 1, wherein the data representative of the transition pattern includes a model of the transition pattern.

9. Software, that is computer readable instructions recorded on computer readable media that when executed by a computer causes it to operate in accordance with the method according to claim 1.

10. A computer system for generating verifying data of a user, the system comprising:

sound capture device to receive time series voice data from the user responding to a sequence of questions multiple levels of difficulty;

a voice assessment module to determine a transition pattern in the voice data based on a plurality of comparisons of subsets of the voice data corresponding to responses to questions having one or more different levels of difficulty; and

a data store to store a representation of the transition pattern for use in future verification of user.

11. A computer implemented method for verifying a user as having previously responded to a first sequence of questions, the method comprising:

receiving time series voice data from the user responding to a second sequence of having multiple levels of difficulty;

determining a transition pattern in the voice data based on a plurality of comparisons of subsets of the voice data corresponding to responses to questions having one or more different levels of difficulty; and

verifying the user based on a comparison of the transition pattern to a previously stored representation of a transition pattern determined from the responses to the first sequence of questions.

12. The computer implemented method according to claim 11, wherein the step of verifying the user comprises determining a likelihood score that that the user previously responded to the first sequence of questions.

13. The computer implemented method according to claim 11, wherein the method further comprises determining the previously stored representation of the transition pattern by:

receiving time series voice data from the user responding to the first sequence of questions having multiple levels of difficulty;

determining a transition pattern in the voice data based on a plurality of comparisons of subsets of the voice data corresponding to responses to questions having one or more different levels of difficulty; and

storing a representation of the transition pattern for use as the previously stored representation of the transition pattern.

14. The computer implemented method according to claim 11, wherein the first sequence of questions is an unproctered test to test the user for suitability or qualification to a role the user is being tested for.

15. The computer implemented method according to claim 13, wherein the method of determining the transition pattern comprises determining from each subset one or more of the following features:

pitch,

first formant,

second formant,

third formant,

glottal excitation, and

place of articulation.

16. The computer implemented method according to claim 11, wherein the comparison of subsets is based on a comparison of one or more features of the subsets, and the method comprises determining a probability distribution function of one or more features of each subset.

17. The computer implemented method according to claim 16, wherein the comparison of subsets is based on a distance measure of the respective probability distribution function.

18. The computer implemented method according to claim 17, wherein the questions are in substantially increasing or decreasing order of difficulty such that the distance measures are combined to form a time series vector.

19. The computer implemented method according to claim 11, wherein the data representative of the transition pattern includes a model of the transition pattern.

20. The computer implemented method according to claim 11 wherein the method further comprises receiving further data from the user while responding to the second sequence of questions being one or more of:

data representative of mouse, stylus or other pointing device movements, including pressure information,

data representative of alphanumeric key input; and

data representative of biological features of the user.

21. The computer implemented method according to claim 11, wherein the comparison of the transition pattern to the previously stored representation of a transition pattern is based on an assessment of whether the transition patterns share a similar statistical distribution.

22. Software, that is computer readable instructions recorded on computer readable media that when executed by a computer causes it to operate in accordance with the method according to claim 11.

23. A computer system to verify a user as having previously performed a first cognitive ability test, the system comprising:

sound capture device to receive time series voice data from the user responding to a sequence of questions that comprise a second cognitive ability test, wherein the questions expose the user to different levels of cognitive load;

a voice assessment module to determine a transition pattern in the voice data based on a plurality of comparisons of subsets of the voice data corresponding to responses to questions having one or more different levels of difficulty; and

a verification module to verify the user based on a comparison of the transition pattern to a previously stored representation of a transition pattern determined from the responses to the first sequence of questions.