Speech signal processing for determining psychological or physiological characteristics using a knowledge base

• US 6,006,188 A
• Filed: 03/19/1997
• Issued: 12/21/1999
• Est. Priority Date: 03/19/1997
• Status: Expired due to Fees

• Current Assignee: Dendrite, Inc.
• Original Assignee: Dendrite, Inc.
• Inventors: Baker, George, Yarigin, Vitaly, Bogdashevsky, Rostislav, Alexeev, Vladimir, Stanton, Harrison
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Smits, Talivaldis Ivars
• Application Number:
  US08/820,566
• Time in Patent Office: 1,007 Days
• Field of Search: 704/270, 434/236, 434/262, 706/50
• US Class Current: 704/270
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

Abstract

A speech-based system for assessing the psychological, physiological, or other characteristics of a test subject is described. The system includes a knowledge base that stores one or more speech models, where each speech model corresponds to a characteristic of a group of reference subjects. Signal processing circuitry, which may be implemented in hardware, software and/or firmware, compares the test speech parameters of a test subject with the speech models. In one embodiment, each speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects. The speech model is independent of a priori knowledge of style parameters associated with the voice or speech. The system includes speech parameterization circuitry for generating the test parameters in response to the test subject'"'"'s speech. This circuitry includes speech acquisition circuitry, which may be located remotely from the knowledge base. The system further includes output circuitry for outputting at least one indicator of a characteristic in response to the comparison performed by the signal processing circuitry. The characteristic may be time-varying, in which case the output circuitry outputs the characteristic in a time-varying manner. The output circuitry also may output a ranking of each output characteristic. In one embodiment, one or more characteristics may indicate the degree of sincerity of the test subject, where the degree of sincerity may vary with time. The system may also be employed to determine the effectiveness of treatment for a psychological or physiological disorder by comparing psychological or physiological characteristics, respectively, before and after treatment.

Claims

1. A psychological assessment system comprising:

a knowledge base including at least one speech model corresponding to at least one psychological characteristic of a plurality of reference subjects, wherein the speech model is independent of a priori knowledge of style parameters; and

signal processing circuitry for comparing the at least one speech model with test speech parameters of a test subject.

2. The system of claim 1, wherein the at least one speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects.

3. The system of claim 1, further comprising:

speech parameterization circuitry for generating the test parameters in response to the test subject'"'"'s speech.

4. The system of claim 3, wherein the speech parameterization circuitry includes speech acquisition circuitry that is remote from the knowledge base.

5. The system of claim 1 further comprising:

output circuitry for outputting at least one indicator of a psychological characteristic in response to the comparison.

6. The system of claim 5, wherein the psychological characteristic is time-varying, and the output circuitry outputs the indicator of the psychological characteristic in a time-varying manner.

7. The system of claim 5, wherein the output circuitry further outputs a ranking of each output psychological characteristic.

8. The system of claim 1, wherein the at least one psychological characteristic indicates degree of sincerity.

9. The system of claim 8, wherein the degree of sincerity varies with time.

10. The system of claim 1, wherein the signal processing circuitry compares psychological characteristics before and after treatment for a psychological disorder, wherein the compared psychological characteristics are generated by the comparison of the at least one speech model with the test speech parameters.

11. A method for psychological assessment comprising the steps of:

providing a knowledge base including at least one speech model corresponding to at least one psychological characteristic of a plurality of reference subjects, wherein the speech model is independent of a priori knowledge of style parameters; and

comparing the at least one speech model with test speech parameters of a test subject.

12. The method of claim 11, wherein the at least one speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects.

13. The method of claim 11, further comprising the step of:

generating the test parameters in response to the test subject'"'"'s speech.

14. The method of claim 13, further comprising the step of acquiring the test subject'"'"'s speech remotely from the knowledge base.

15. The method of claim 11 further comprising the step of:

outputting at least one indicator of a psychological characteristic in response to the comparison.

16. The method of claim 15, wherein the psychological characteristic is time-varying, further comprising the step of outputting the indicator of the psychological characteristic in a time-varying manner.

17. The method of claim 15, the outputting step further comprising the step of outputting a ranking of each output psychological characteristic.

18. The method of claim 11, wherein the at least one psychological characteristic indicates degree of sincerity.

19. The method of claim 18, wherein the degree of sincerity varies with time.

20. The method of claim 11, further comprising the step of comparing psychological characteristics before and after treatment for a psychological disorder, wherein the compared psychological characteristics are generated by the comparison of the at least one speech model with the test speech parameters.

21. A physiological assessment system comprising:

a knowledge base including at least one speech model corresponding to at least one physiological characteristic of a plurality of reference subjects, wherein the speech model includes information concerning the vocal tracts of the reference subjects; and

signal processing circuitry for comparing the at least one speech model with test speech parameters of a test subject.

22. The system of claim 21, wherein the at least one speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects.

23. The system of claim 21, wherein the at least one speech model is independent of a priori knowledge of style parameters.

24. The system of claim 21, further comprising:

speech parameterization circuitry for generating the test parameters in response to the test subject'"'"'s speech.

25. The system of claim 24, wherein the speech parameterization circuitry includes speech acquisition circuitry that is remote from the knowledge base.

26. The system of claim 21 further comprising:

output circuitry for outputting at least one indicator of a physiological characteristic in response to the comparison.

27. The system of claim 26, wherein the physiological characteristic is time-varying, and the output circuitry outputs the indicator of the physiological characteristic in a time-varying manner.

28. The system of claim 26, wherein the output circuitry further outputs a ranking of each output physiological characteristic.

29. The system of claim 21, wherein the signal processing circuitry compares physiological characteristics before and after treatment for a physiological disorder, wherein the compared psychological characteristics are generated by the comparison of the at least one speech model with the test speech parameters.

30. A method for physiological assessment comprising the steps of:

providing a knowledge base including at least one speech model corresponding to at least one physiological characteristic of a plurality of reference subjects, wherein the speech model includes information concerning the vocal tracts of the reference subjects; and

comparing the at least one speech model with test speech parameters of a test subject.

31. The method of claim 30, wherein the at least one speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects.

32. The method of claim 30, wherein the speech model is independent of a priori knowledge of style parameters.

33. The method of claim 30, further comprising the step of:

generating the test parameters in response to the test subject'"'"'s speech.

34. The method of claim 33, further comprising the step of acquiring the test subject'"'"'s speech remotely from the knowledge base.

35. The method of claim 30 further comprising the step of:

outputting at least one indicator of a physiological characteristic in response to the comparison.

36. The method of claim 35, wherein the physiological characteristic is time-varying, further comprising the step of outputting the indicator of the physiological characteristic in a time-varying manner.

37. The method of claim 35, the outputting step further comprising the step of outputting a ranking of each output physiological characteristic.

38. The method of claim 30, further comprising the step of comparing physiological characteristics before and after treatment for a physiological disorder, wherein the compared physiological characteristics are generated by the comparison of the at least one speech model with the test speech parameters.

39. In a system for assessing at least one psychological or physiological characteristic of a test subject, a knowledge base comprising:

at least one speech model corresponding to at least one characteristic, wherein the at least one speech model is independent of a priori knowledge of style parameters, and

a statistical time-ordered series of frequency representations of the speech of a plurality of reference subjects within the at least one speech model.

40. The knowledge base of claim 39, wherein the at least one speech model includes information concerning the vocal tracts of the reference subjects.

41. In a method for assessing at least one psychological or physiological characteristic of a test subject, a method for creating a knowledge base comprising the steps of:

forming at least one speech model corresponding to at least one characteristic, wherein the at least one speech model is independent of a priori knowledge of style parameters, and

generating a statistical time-ordered series of frequency representations of the speech of a plurality of reference subjects within the at least one speech model.

42. The method of claim 41, wherein the at least one speech model includes information concerning the vocal tracts of the reference subjects.

43. A physiological assessment system comprising:

a knowledge base including at least one speech model corresponding to at least one physiological characteristic of a plurality of reference subjects, wherein the at least one speech model is independent of a priori knowledge of style parameters; and

signal processing circuitry for comparing the at least one speech model with test speech parameters of a test subject.

44. The system of claim 43, wherein the at least one speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects.

45. The system of claim 43, further comprising:

speech parameterization circuitry for generating the test parameters in response to the test subject'"'"'s speech.

46. The system of claim 45, wherein the speech parameterization circuitry includes speech acquisition circuitry that is remote from the knowledge base.

47. The system of claim 43, further comprising:

output circuitry for outputting at least one indicator of a physiological characteristic in response to the comparison.

48. The system of claim 47, wherein the physiological characteristic is time-varying, and the output circuitry outputs the indicator of the physiological characteristic in a time-varying manner.

49. The system of claim 47, wherein the output circuitry further outputs a ranking of each output physiological characteristic.

50. The system of claim 43, wherein the signal processing circuitry compares physiological characteristics before and after treatment for a physiological disorder, wherein the compared psychological characteristics are generated by the comparison of the at least one speech model with the test speech parameters.

51. A method for physiological assessment comprising the steps of:

providing a knowledge base including at least one speech model corresponding to at least one physiological characteristic of a plurality of reference subjects, wherein the at least one speech model is independent of a priori knowledge of style parameters; and

comparing the at least one speech model with test speech parameters of a test subject.

52. The method of claim 51, wherein the at least one speech model is represented by a statistical time-ordered series of frequency representations of the speech of the reference subjects.

53. The method of claim 51, further comprising the step of:

generating the test parameters in response to the test subject'"'"'s speech.

54. The method of claim 53, further comprising the step of acquiring the test subject'"'"'s speech remotely from the knowledge base.

55. The method of claim 51, further comprising the step of:

outputting at least one indicator of a physiological characteristic in response to the comparison.

56. The method of claim 55, wherein the physiological characteristic is time-varying, further comprising the step of outputting the indicator of the physiological characteristic in a time-varying manner.

57. The method of claim 55, the outputting step further comprising the step of outputting a ranking of each output physiological characteristic.

58. The method of claim 51, further comprising the step of comparing physiological characteristics before and after treatment for a physiological disorder, wherein the compared physiological characteristics are generated by the comparison of the at least one speech model with the test speech parameters.

59. In a system for assessing at least one physiological characteristic of a test subject, a knowledge base comprising:

at least one speech model corresponding to each characteristic, wherein the speech model includes information concerning the vocal tracts of the reference subjects; and

a statistical time-ordered series of frequency representations of the speech of a plurality of reference subjects within the at least one speech model.

60. A system for assessing at least one psychological or physiological characteristic of a test subject, the system comprising:

speech parameterization circuitry for generating test speech parameters in response to the test subject'"'"'s speech; and

signal processing circuitry for comparing the test speech parameters with at least one speech model from a knowledge base,

wherein the at least one speech model corresponds to at least one psychological or physiological characteristic of a plurality of reference subjects, and the at least one speech model is independent of a priori knowledge of style parameters.

61. A system for assessing at least one physiological characteristic of a test subject, the system comprising:

speech parameterization circuitry for generating test speech parameters in response to the test subject'"'"'s speech; and

signal processing circuitry for comparing the test speech parameters of with at least one speech model from a knowledge base,

wherein the at least one speech model corresponds to at least one physiological characteristic of a plurality of reference subjects, and includes information concerning the vocal tracts of the reference subjects.

62. A method for assessing at least one psychological or physiological characteristic of a test subject, the method comprising the steps of:

generating test speech parameters in response to the test subject'"'"'s speech; and

comparing the test speech parameters with at least one speech model from a knowledge base,

wherein the at least one speech model corresponds to at least one psychological or physiological characteristic of a plurality of reference subjects, and the at least one speech model is independent of a priori knowledge of style parameters.

63. A method for assessing at least one physiological characteristic of a test subject, the method comprising the steps of:

generating test speech parameters in response to the test subject'"'"'s speech; and

comparing the test speech parameters with at least one speech model from a knowledge base,

wherein the at least one speech model corresponds to at least one physiological characteristic of a plurality of reference subjects, and includes information concerning the vocal tracts of the reference subjects.