METHOD AND APPARATUS FOR DETERMINING EMOTION INFORMATION FROM USER VOICE

• US 20150310878A1
• Filed: 04/27/2015
• Published: 10/29/2015
• Est. Priority Date: 04/25/2014
• Status: Abandoned Application

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: BRONAKOWSKI, Lukasz Jakub, STASZUK, Arleta, TKACZUK, Jakub
• Application Number:
  US14/696,649
• Publication Number:
  US 20150310878A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/63 for estimating an emotional state
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals
• CPC Class Codes: G10L 25/93 Discriminating between voiced and unvoiced parts of speech signals G10L25/90 takes precedence

Abstract

A method of determining emotion information from a voice is provided. The method includes receiving a voice frame obtained by converting a sound generated by a user into an electrical signal, detecting phonation information and articulation information, the phonation information being related to phonation of the user and the articulation information being related to articulation of the user, from the voice frame, and determining user emotion information corresponding to the phonation information and the articulation information.

Claims

1. A method of determining emotion information from a voice, the method comprising:

receiving a voice frame obtained by converting a sound generated by a user into an electrical signal;

detecting phonation information and articulation information, the phonation information being related to phonation of the user and the articulation information being related to articulation of the user, from the voice frame; and

determining user emotion information corresponding to the phonation information and the articulation information.

2. The method of claim 1, wherein the phonation information includes information related to glottides of the user.

3. The method of claim 1, wherein the phonation information includes at least one of information about a size of a vocal cord of the user, information about braking power of tissues of the vocal cord of the user, and information about an elastic force of the tissues of the vocal cord of the user.

4. The method of claim 1, wherein the phonation information includes a fundamental frequency of the voice frame.

5. The method of claim 1, wherein the articulation information includes information related to a vocal tract of the user.

6. The method of claim 1, wherein the articulation information includes a sound characteristic of the voice frame.

7. The method of claim 1, wherein the detecting of the phonation information and the articulation information comprises detecting information related to a level of tension of glottides of the user.

8. The method of claim 7, wherein the detecting of the information related to the level of tension of the glottides comprises:

filtering noise except for a fundamental frequency of the voice frame; and

filtering a band of a voiceless sound.

9. The method of claim 7, wherein the detecting of the information related to the level of tension of the glottides includes:

generating a divided frame by dividing the voice frame by a time unit;

determining energy of the divided frame;

determining a ratio of parts of the divided frame that have an energy level equal to or greater than a first threshold value; and

detecting information related to the level of tension of the glottides of the user from a voice frame in which the determined ratio exceeds a second threshold value.

10. The method of claim 1, further comprising:

determining a gender of the user by using at least one piece of information corresponding to the phonation information and the articulation information,

wherein the determining of the user emotion information includes determining the user emotion information by using the at least one piece of information corresponding to the phonation information and the articulation information.

11. The method of claim 1, wherein the detecting of the phonation information and the articulation information includes dividing the voice frame by a time unit.

12. An electronic apparatus comprising:

a microphone configured to convert an input voice signal into an electrical signal;

a speaker configured to output the electrical signal;

a screen configured to display information; and

at least one controller configured to process a program for determining user emotion information,

wherein the program for determining the user emotion information includes commands for;

converting the electrical signal into a voice frame,

detecting phonation information and articulation information, the phonation information being related to phonation of the user and the articulation information being related to articulation of the user, from the voice frame, and

determining the user emotion information corresponding to the phonation information and the articulation information.

13. The electronic apparatus of claim 12, wherein the phonation information includes information related to glottides of the user.

14. The electronic apparatus of claim 13, wherein the phonation information includes at least one of information about a size of a vocal cord of the user, information about braking power of tissues of the vocal cord of the user, and information about an elastic force of the tissues of the vocal cord of the user.

15. The electronic apparatus of claim 12, wherein the articulation information includes information related to a vocal tract of the user.

16. The electronic apparatus of claim 12, wherein the program for determining the user emotion information further includes commands for:

filtering noise except for a fundamental frequency of the voice frame, and

filtering a band of a voiceless sound.

17. The electronic apparatus of claim 12, wherein the program for determining the user emotion information further includes commands for:

generating a divided frame by dividing the voice frame by a time unit,

determining a ratio of parts of the divided frame that have an energy level equal to or greater than a first threshold value, and

detecting information related to the level of tension of glottides of the user from a voice frame in which the determined ratio exceeds a second threshold value.

18. The electronic apparatus of claim 12, further comprising a storage unit configured to store a database, which includes the phonation information, the articulation information, and the user emotion information corresponding to the phonation information and the articulation information.

19. The electronic apparatus of claim 12, wherein the program for determining the user emotion information further includes commands for:

determining a gender of the user by using at least one piece of information corresponding to the phonation information and the articulation information, and

determining the user emotion information by using the at least one piece of information corresponding to the phonation information and the articulation information.

20. The electronic apparatus of claim 12, further comprising a storage unit configured to store a first database including emotion information about a first gender corresponding to the phonation information and the articulation information, and to store a second database including emotion information about a second gender corresponding to the phonation information and the articulation information.