System method and article of manufacture for detecting emotion in voice signals by utilizing statistics for voice signal parameters

• US 6,275,806 B1
• Filed: 08/31/1999
• Issued: 08/14/2001
• Est. Priority Date: 08/31/1999
• Status: Expired due to Term

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Andersen Consulting (Accenture PLC)
• Inventors: Pertrushin, Valery A.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Armstrong, Angela A.
• Application Number:
  US09/388,909
• Time in Patent Office: 714 Days
• Field of Search: 704/270, 704/272, 704/275, 704/200, 704/240, 434/185
• US Class Current: 704/272
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

Abstract

A database is provided. The database includes statistics of human associations of human voice parameters with emotions. A voice signal is received. At least one feature of this voice signal is extracted. This extracted voice feature is then compared to the voice parameters in the database. An emotion is selected from the database based on the comparison of the extracted voice feature to the voice parameters. Input from the user is received. This input includes a user-determined emotion. The user-determined emotion is compared with the emotion selected from the database. The selected emotion is output and a determination as to whether the user-determined emotion matches the emotion selected from the database is made. A prize is then awarded to the user if the user-determined emotion is determined to match the selected emotion from the database.

Claims

1. A method for ascertaining an emotion in a voice by utilizing statistics comprising the steps of:

(a) providing a database having statistics including statistics of human associations of voice parameters with emotions;

(b) receiving a voice signal;

(c) extracting at least one feature of the voice signal;

(d) comparing the extracted voice feature to the voice parameters in the database;

(e) selecting an emotion from the database based on the comparison of the extracted voice feature to the voice parameters;

(f) receiving an input from the user, wherein the input includes a user-determined emotion;

(g) comparing the user-determined emotion with the emotion selected from the database;

(h) outputting the selected emotion;

(i) determining whether the user-determined emotion matches the emotion selected from the database; and

(j) awarding a prize to the user if the user-determined emotion matches the selected emotion from the database.

2. A method as recited in claim 1, wherein the database includes probabilities of particular voice features being associated with an emotion.

3. A method as recited in claim 2, wherein the selection of the emotion from the database includes analyzing the probabilities and selecting the most probable emotion based on the probabilities.

4. A method as recited in claim 2, wherein the probabilities of the database include performance confusion statistics.

5. A method as recited in claim 1, wherein the statistics in the database include self-recognition statistics.

6. A method as recited in claim 1, wherein a plurality of features are extracted including the maximum value of the fundamental frequency, the standard deviation of the fundamental frequency, the range of the fundamental frequency, the mean of the fundamental frequency, the mean of the bandwidth of the first formant, the mean of the bandwidth of the second formant, the standard deviation of energy, the speaking rate, the slope of the fundamental frequency, the maximum value of the first formant, the maximum value of the energy, the range of the energy, the range of the second formant, and the range of the first formant.

7. A method as recited in claim 1, wherein the user is diagnosed as being autistic.

8. A method as recited in claim 1, wherein the user competes against at least one of:

another user and a computer.

9. A computer program embodied on a computer readable medium for detecting emotion in a voice by utilizing statistics comprising:

(a) a code segment that provides a database having statistics including statistics of human associations of voice parameters with emotions;

(b) a code segment that receives a voice signal;

(c) a code segment that extracts at least one feature of the voice signal;

(d) a code segment that compares the extracted voice feature to the voice parameters in the database;

(e) a code segment that selects an emotion from the database based on the comparison of the extracted voice feature to the voice parameters;

(f) a code segment that receives an input from the user, wherein the input includes a user-determined emotion;

(g) a code segment that compares the user-determined emotion with the emotion selected from the database;

(h) a code segment that outputs the selected emotion;

(i) a code segment that determines whether the user-determined emotion matches the emotion selected from the database; and

a code segment that awards a prize to the user if the user-determined emotion matches the selected emotion from the database.

10. A computer program as recited in claim 9, wherein the database includes probabilities of particular voice features being associated with an emotion.

11. A computer program as recited in claim 10, wherein the selection of the emotion from the database includes analyzing the probabilities and selecting the most probable emotion based on the probabilities.

12. A computer program as recited in claim 10, wherein the probabilities of the database include performance confusion statistics.

13. A computer program as recited in claim 9, wherein the statistics in the database include self-recognition statistics.

14. A computer program as recited in claim 9, wherein a plurality of features are extracted including the maximum value of the fundamental frequency, the standard deviation of the fundamental frequency, the range of the fundamental frequency, the mean of the fundamental frequency, the mean of the bandwidth of the first formant, the mean of the bandwidth of the second formant, the standard deviation of energy, the speaking rate, the slope of the fundamental frequency, the maximum value of the first formant, the maximum value of the energy, the range of the energy, the range of the second formant, and the range of the first formant.

15. A system for detecting emotion in a voice by utilizing statistics comprising:

(a) logic that provides a database having statistics including statistics of human associations of voice parameters with emotions;

(b) logic that receives a voice signal;

(c) logic that extracts at least one feature of the voice signal;

(d) logic that compares the extracted voice feature to the voice parameters in the database;

(e) logic that selects an emotion from the database based on the comparison of the extracted voice feature to the voice parameters;

(f) logic that receives an input from the user, wherein the input includes a user-determined emotion;

(g) logic that compares the user-determined emotion with the emotion selected from the database;

(h) logic that outputs selected emotion;

(i) logic that determines whether the user-determined emotion matches the emotion selected from the database; and

(j) logic that awards a prize to the user if the user-determined emotion matches the selected emotion from the database.

16. A system as recited in claim 15, wherein the database includes probabilities of particular voice features being associated with an emotion.

17. A system as recited in claim 16, wherein the selection of the emotion from the database includes analyzing the probabilities and selecting the most probable emotion based on the probabilities.

18. A system as recited in claim 16, wherein the probabilities of the database include performance confusion statistics.

19. A system as recited in claim 15, wherein the statistics in the database include self-recognition statistics.

20. A system as recited in claim 15, wherein a plurality of features are extracted including the maximum value of the fundamental frequency, the standard deviation of the fundamental frequency, the range of the fundamental frequency, the mean of the fundamental frequency, the mean of the bandwidth of the first formant, the mean of the bandwidth of the second formant, the standard deviation of energy, the speaking rate, the slope of the fundamental frequency, the maximum value of the first formant, the maximum value of the energy, the range of the energy, the range of the second formant, and the range of the first formant.