Method for quantifying psychological stress levels using voice pattern samples

• US 7,321,855 B2
• Filed: 12/15/2003
• Issued: 01/22/2008
• Est. Priority Date: 12/15/2003
• Status: Active Grant

• Current Assignee: Charles Humble
• Original Assignee: Charles Humble
• Inventors: Humble, Charles
• Primary Examiner(s): Neway; Samuel G
• Application Number:
  US10/737,530
• Publication Number:
  US 20050131692A1
• Time in Patent Office: 1,499 Days
• Field of Search: 704/273
• 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 computer-implemented method of assigning a numeric score to a voice pattern sample of a human subject wherein the score is indicative of the psychological stress level of the human subject. A verbal utterance of a human subject is converted into electrical signals to provide a subject wave pattern. The pattern is quantified and compared with known voice pattern characteristics which exhibit a sequential progression in the degree of blocking in the pattern, wherein each of the known voice patterns is assigned a numerical value range. A numerical value obtained from iterative calculations is assigned to the subject wave pattern based on the comparison. The numerical value represents the degree of blocking present in the subject wave pattern which correlates to the amount of psychological stress exhibited by the human subject.

Claims

1. A computer-implemented method of assigning a numeric score to a voice pattern sample of a human subject wherein the score is indicative of the psychological stress level of the human subject, comprising the steps of:

converting the verbal utterance of a human subject into electrical signals plotted on a Cartesian coordinate system to provide a subject wave pattern having x and y axes;

filtering the subject wave pattern to remove extraneous values at the leading and trailing edges of the wave pattern to provide a Relevant Pattern Location for analysis;

providing a consecutive series of known voice pattern characteristics, wherein the consecutive series exhibits a sequential progression in the degree of blocking in the pattern;

assigning a numerical value range to each of the consecutive series of known voice pattern characteristics based on the degree of blocking in the pattern;

analyzing the subject wave pattern within the Relevant Pattern Location to provide quantified data defining a set of characteristics for the subject wave pattern;

providing quantified data defining a set of characteristics for the consecutive series of known voice pattern characteristics; and

comparing the quantified data of subject wave pattern within the Relevant Pattern Location to quantified data for each of the known voice pattern characteristics to ascertain a match between the subject wave pattern and one or more of the known voice pattern characteristics; and

returning a numerical value for the subject wave pattern based on the one or more matching known voice pattern characteristics.

2. The method of claim 1, further comprising the step of:

applying a value-averaging algorithm to the y component of the subject wave pattern to refine the subject wave pattern.

3. The method of claim 1, further comprising the step of measuring the subject wave pattern to determine if the subject wave pattern conforms to required characteristics for optimal analysis.

4. The method of claim 3, wherein the step of measuring the subject wave pattern further comprises determining the width of the pattern relevant to the x-axis.

5. The method of claim 4, further comprising the step of returning an error message if the width of the pattern is greater than 6000 integers.

6. The method of claim 1, wherein the quantified data for the subject wave pattern and the known voice pattern characteristics includes the wave frequency.

7. The method of claim 1, wherein the quantified data for the subject wave pattern and the known voice pattern characteristics includes the wave amplitude.

8. The method of claim 1, wherein the quantified data for the subject wave pattern and the known voice pattern characteristics includes the number of cycles in refined subject wave pattern.

9. The method of claim 8, further comprising the step of returning an error message if the number of cycles is less than 6 cycles.

10. The method of claim 1, further comprising the steps of:

filtering the subject wave pattern to remove values at the leading edge of the wave pattern where y=0; and

filtering the subject wave pattern to remove values at the trailing edge of the wave pattern where y=0.

11. The method of claim 1, wherein the step of analyzing the subject wave pattern further includes the step of forming an maximum point array based on the peak of each cycle in the subject wave pattern.

12. The method of claim 1, wherein the step of analyzing the subject wave pattern further includes the step of forming an minimum point array based on the base of each cycle in the subject wave pattern.

13. The method of claim 1, wherein the step of analyzing the subject wave pattern further includes forming an midpoint point array based on the midpoint between each maximum point and minimum point in the subject wave pattern.

14. A computer-implemented method of assigning a numeric score to a voice pattern sample of a human subject wherein the score is indicative of the psychological stress level of the human subject, comprising the steps of:

converting the verbal utterance of a human subject into electrical signals plotted on a Cartesian coordinate system to provide a subject wave pattern;

measuring the subject wave pattern to determine if the subject wave pattern conforms to required characteristics for optimal analysis;

filtering the subject wave pattern to remove extraneous values at the leading and trailing edges the wave pattern to provide a Relevant Pattern Location for analysis;

forming a midpoint point array based on the midpoint of each half cycle in the subject wave pattern;

detecting the largest segment of the midpoint array where y is approximately constant to provide an Isolated Pattern Location;

enlarging the Isolated Pattern Location relevant to the x-axis to encompass approximately 50% of the Relevant Pattern Location;

analyzing the wave pattern within the Isolated Pattern Location to provide quantified data defining a set of characteristics for the wave pattern;

providing a consecutive series of known voice pattern characteristics, wherein the consecutive series exhibits a sequential progression in the degree of blocking in the pattern;

assigning a numerical value range to each of the consecutive series of known voice pattern characteristics based on the degree of blocking in the pattern;

comparing the quantified data for the wave pattern within the Isolated Pattern Location to the parameters of known voice pattern characteristics to ascertain a match between wave pattern in the Isolated Pattern Location and one or more of the known voice pattern characteristics; and

returning a numerical value for the Isolated Pattern Location based on the matching of parameters with known voice pattern characteristics.

15. The method of claim 14, further comprising the step of applying a value-averaging algorithm to the y-component of the subject wave pattern within the refined subject wave pattern.

16. The method of claim 14, wherein the step of analyzing the wave pattern within the Isolated Pattern Location further includes calculating the amplitude of the wave pattern.

17. The method of claim 14, wherein the step of analyzing the wave pattern within the Isolated Pattern Location further includes calculating the percentage of the Isolated Pattern Location relevant to the Relevant Pattern Location.

18. The method of claim 14, wherein the step of analyzing the wave pattern within the Isolated Pattern Location further includes calculating the slope of the wave pattern.

19. The method of claim 14, wherein the step of quantifying the wave pattern within the Isolated Pattern Location further includes calculating the percentage of the area under the wave pattern within the Isolated Pattern Location relevant to a projected rectangular area having the minimal dimensions to encompass the Isolated Pattern Location.