Method and system for verifying and enabling user access based on voice parameters

• US 7,054,811 B2
• Filed: 10/06/2004
• Issued: 05/30/2006
• Est. Priority Date: 11/06/2002
• Status: Expired due to Fees

• Current Assignee: Cellmax Systems Limited
• Original Assignee: Cellmax Systems Limited
• Inventors: Barzilay, Ziv
• Primary Examiner(s): Chawan, Vijay
• Assistant Examiner(s): SHORTLEDGE, THOMAS E
• Application Number:
  US10/958,498
• Publication Number:
  US 20050096906A1
• Time in Patent Office: 601 Days
• Field of Search: 704/231, 704/246, 704/270.1, 704/270, 704/275, 704/273, 704/247
• US Class Current: 704/246
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/105 Financial transactions and auctions, e.g. bidding auctioneering devices in packet switching networks H04L12/1804
• CPC Class Codes: H04M 2203/1058 Shopping and product ordering
• CPC Class Codes: H04M 2207/18 wireless networks
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 7/006 Networks other than PSTN/ISDN providing telephone service, e.g. Voice over Internet Protocol (VoIP) , including next generation networks with a packet-switched transport layer H04L65/00 takes precedence; aspects not specific to the type of network H04M3/00; special services in those networks H04M3/42

Abstract

A system for verifying and enabling user access, which includes a voice registration unit for providing a substantially unique and initial identification of each of a plurality of the speaker/users by finding the speaker/user'"'"'s voice parameters in a voice registration sample and storing same in a database. The system also includes a voice authenticating unit for substantially absolute verification of an identity of one of said plurality of users. The voice authenticating unit includes a recognition unit for providing a voice authentication sample, and being operative with the database. The voice authenticating unit also includes a decision unit operative with the recognition unit and the database to decide whether the user is the same as the person of the same identity registered with the system, such that the identity of one of the plurality of users is substantially absolutely verified.

Claims

1. A system for verifying and enabling user access based on voice parameters, said system comprising:

a voice registration unit for registering a user by finding the user'"'"'s voice parameters in a voice registration sample containing raw data and storing same in a database to provide a substantially unique and initial identification of each of a plurality of users; and

a voice authenticating unit for substantially accurate verification of an identity of one of said plurality of users, said voice authenticating unit comprising;

a recognition unit for providing a voice authentication sample, and being operative with said database; and

a decision unit operative with said recognition unit and said database to decide whether the user associated with said voice authentication sample is the same as the identity of the user registered with the system and associated with said voice registration sample,

such that said identity of one of said plurality of users is substantially accurately verified for access purposes,

wherein the user'"'"'s voice parameters are determined in accordance with a method comprising;

eliciting a voice registration sample of at least one utterance from the user, wherein the system asks the user to say some of his characteristics;

fractally analyzing said sample of at least one utterance, wherein the raw data of the user'"'"'s voice is investigated and each sample gives a set of non-dimensional numbers, which characterize said voice registration sample uniquely;

spectrographically analyzing said sample of at least one utterance, wherein said raw data is investigated and generates a unique voiceprint for purposes of substantially unique identification, thereby allowing extraction of geometrical characteristics relative to the user'"'"'s voice; and

determining the Lyapunov exponents of said sample of at least one utterance, wherein a spectrum of said exponents are computed, which further characterize said voice registration sample uniquely,

wherein the identification is done in stages, by eliciting a voice authentication sample of at least one utterance from the user, and repeating said fractal analysis, spectrographic analysis and Lyapunov exponent determination steps, while automatically adapting to a continuous change of the voice parameters to overcome inherent instability associated therewith, by adding to the voice registration sample with each instance of successful voice authentication,

and wherein the identification process provides said substantially accurate verification by applying a branch and bound algorithm to the voice parameters of said voice registration sample to determine the closest fit with the voice parameters of said voice authentication sample,

such that the system proceeds efficiently and within a practical time frame to shorten calculation periods and achieve a calculation result in a manageable fashion, in order to decide whether the user calling in for authentication is the same as the person of that identity registered in the system.

2. The system of claim 1, wherein the system is applied to e-commerce transactions over a communications network, wherein said voice registration sample is provided in a phone call.

3. The system of claim 2, wherein the communications network is a land-based line.

4. The system of claim 2, wherein the communications network is a wireless network.

5. The system of claim 2, wherein the communications network is a voice over Internet protocol network.

6. The system of claim 1, wherein the identity of one of said plurality of users during said call is verified to a substantially unambiguous accuracy.

7. The system of claim 1, wherein the system is applied to banking transactions.

8. The system of claim 7, wherein the system is global and a banking transaction can be made at any time, and wherein check writing is replaced by a voice transaction.

9. The system of claim 7, wherein ATM transactions are voice implemented.

10. The system of claim 7, wherein said voice authentication sample is used to make said access decision.

11. The system of claim 1, wherein the system is applied to credit transactions.

12. The system of claim 11, wherein said voice authentication sample replaces the magnetic strip on a credit card.

13. The system of claim 11, wherein said voice authentication sample replaces the caller ID feature in a telephone system.

14. The system of claim 11, wherein said voice authentication sample replaces the leaving of one'"'"'s name in a voice message.

15. The system of claim 11, wherein the system prevents fraud and replaces plastic credit cards by implementation of a simple voice protocol.

16. The system of claim 1, wherein the system is applied to employee identification.

17. The system of claim 1, wherein the system is applied to airport security for traveler identification and a central information center knows what terminal the traveler is in now and whether he is allowed to board a specific flight.

18. The system of claim 1, wherein the system is applied to automotive keyless entry for prevention of auto theft, by replacing key entry and remote beeper entry by voice programmed entry.

19. The system of claim 1, wherein the system is applied to access control, operated and controlled by authorized voice only.

20. The system of claim 19, wherein the access is applied to at least one of the following:

a safe;

a vault;

a safety deposit box; and

a voice control service.

21. The system of claim 1, wherein the system is applied to law enforcement for identification from voice samples and fraud prevention.

22. The system of claim 1, wherein the system is applied to military for decision-making to control activities based on accurate real-time identification of personnel.

23. The system of claim 1, wherein said voice authentication sample is in any language.

24. The system of claim 1, wherein an attempt is made to disguise said voice authentication sample.

25. The system of claim 1, wherein said voice authentication sample is distorted unintentionally by the user.

26. The system of claim 1, wherein said voice authentication sample is distorted unintentionally by the user because of health problems.

27. The system of claim 1, wherein said system can differentiate between a natural voice authentication sample and an unnatural voice authentication sample.

28. The system of claim 27, wherein said unnatural voice authentication sample is made using a tape recording.

29. The system of claim 1, wherein said voice registration sample is text-dependent.

30. The system of claim 1, wherein said voice registration sample is text-independent.

31. The system of claim 1, wherein said voice authentication sample is text-dependent.

32. The system of claim 1, wherein said voice authentication sample is text-independent.

33. The system of claim 1, further comprising an interactive voice response unit for receiving a voice authentication sample from a user.

34. The system of claim 1, further comprising an intercom for receiving a voice authentication sample from a user.

35. The system of claim 34, wherein said intercom for receiving a voice authentication sample from a user is for purposes of access to a facility.

36. The system of claim 34, wherein said intercom for receiving a voice authentication sample from a user is for purposes of access to his home.

37. The system of claim 34, wherein said intercom for receiving a voice authentication sample from a user is for purposes of access to an automobile.

38. The system of claim 1, wherein said database is a virtual key owner'"'"'s database.

39. The system of claim 1, wherein said database is a passport database.

40. The system of claim 1, wherein said database is a “

wanted list”

database.

41. The system of claim 1, wherein said database is a car owners database.

42. The system of claim 1, wherein said database is a database of automotive drivers.

43. The system of claim 1, wherein said voice authentication sample replaces the key to a car.

44. The system of claim 1, wherein said voice authentication sample replaces the key to a house.

45. The system of claim 1, further comprising at least one voice recording device for receiving a voice authentication sample from a plurality of users in a plurality of locations.

46. The system of claim 45, wherein said plurality of users in a plurality of locations is at an airport.

47. The system of claim 45, wherein said plurality of users in a plurality of locations is at a corporate facility.

48. The system of claim 45, wherein said plurality of users in a plurality of locations is at a government facility.

49. A method for the purpose of interactively conducting secure, computerized, voice-based e-commerce transactions operated over a communications network, by registering and authenticating a plurality of users with a system provided for the purpose, the method comprising;

registering the plurality of users with the system, by having each user call into the system for registration;

eliciting a voice registration sample containing raw data of at least one utterance if from the user, wherein the system asks the user to say some of his characteristics;

fractally analyzing said sample of at least one utterance, wherein the raw data of the speaker'"'"'s voice is investigated and each sample gives a set of non-dimensional numbers, which characterize said voice registration sample uniquely;

spectrographically analyzing said sample of at least one utterance, wherein said raw data is investigated and generates a unique voiceprint for purposes of substantially unique identification, thereby allowing extraction of geometrical characteristics relative to the speaker'"'"'s voice; and

determining the Lyapunov exponents of said sample of at least one utterance, wherein a spectrum of said exponents are computed, which further characterize said voice registration sample uniquely,

wherein the identification is done in stages, by eliciting a voice authentication sample of at least one utterance from the user; and

repeating said fractal analysis, spectrographic analysis and Lyapunov exponent determination steps, while automatically adapting to a continuous change of the voice parameters to overcome inherent instability associated therewith, by adding to the voice registration sample with each instance of successful voice authentication,

and wherein the identification process provides said substantially accurate verification by applying a branch and bound algorithm to the voice parameters of said voice registration sample to determine the closest fit with the voice parameters of said voice authentication sample,

such that the system proceeds efficiently and within a practical time frame to shorten calculation periods and achieve a calculation result in a manageable fashion, in order to decide whether the user calling in for authentication is the same as the person of that identity registered in the system.

50. The method of claim 49, further comprising learning by the system after each authentication of each of said plurality of users.

51. The method of claim 49, further comprising repeating the steps after calling into the system, in the case where the identification fails, thereby allowing the system another chance for identification.

52. The method of claim 49, further comprising analyzing by the technique of dynamic time warping (DTW).

53. The method of claim 49, wherein spectrographically analyzing further comprises using an FFT to provide a sixteen (16) dimension feature vector.

54. The method of claim 49, wherein said branch and bound algorithm comprises a column-wise expansion of a dynamic time warping (DTW) matrix of said at least one utterance.

55. A method for verifying and enabling user access based on voice parameters, by registering and authenticating a plurality of users with a system provided for the purpose, said method comprising:

registering the plurality of users with the system by having each user call into the system for registration;

eliciting a voice registration sample containing raw data of at least one utterance from the user, wherein the system asks the user to say some of his characteristics;

fractally analyzing said sample of at least one utterance, wherein the raw data of the speaker'"'"'s voice is investigated and each sample gives a set of non-dimensional numbers, which characterize said voice registration sample uniquely;

spectrographically analyzing said sample of at least one utterance, wherein said raw data is investigated and generates a unique voiceprint for purposes of substantially unique identification, thereby allowing extraction of geometrical characteristics relative to the speaker'"'"'s voice; and

determining the Lyapunov exponents of said sample of at least one utterance, wherein a spectrum of said exponents are computed, which further characterize said voice registration sample uniquely,

wherein the identification is done in stages, by eliciting a voice authentication sample of at least one utterance from the user; and

repeating said fractal analysis, spectrographic analysis and Lyapunov exponent determination steps, while automatically adapting to a continuous change of the voice parameters to overcome inherent instability associated therewith, by adding to the voice registration sample with each instance of successful voice authentication,

and wherein the identification process provides said substantially accurate verification by applying a branch and bound algorithm to the voice parameters of said voice registration sample to determine the closest fit with the voice parameters of said voice authentication sample,

such that the system proceeds efficiently and within a practical time frame to shorten calculation periods and achieve a calculation result in a manageable fashion, in order to decide whether the user calling in for authentication is the same as the person of that identity registered in the system.