METHOD FOR DETERMINING ALCOHOL CONSUMPTION, AND RECORDING MEDIUM AND TERMINAL FOR CARRYING OUT SAME

• US 20170004848A1
• Filed: 01/24/2014
• Published: 01/05/2017
• Est. Priority Date: 01/24/2014
• Status: Active Grant

• Current Assignee: Foundation of Soongsil University-Industry Cooperation
• Original Assignee: Foundation of Soongsil University-Industry Cooperation
• Inventors: BAE, Myung Jin, LEE, Sang Gil, BAEK, Geum Ran
• Granted Patent: US 9,934,793 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 25/21 the extracted parameters being power information
• CPC Class Codes: G10L 25/48 specially adapted for particular use
• CPC Class Codes: G10L 25/66 for extracting parameters related to health condition detecting or measuring for diagnostic purposes A61B5/00
• CPC Class Codes: G10L 25/84 for discriminating voice from noise

Abstract

Disclosed are a method for determining whether a person is drunk after consuming alcohol capable of analyzing alcohol consumption in a time domain by analyzing a voice, and a recording medium and a terminal for carrying out same. An alcohol consumption-determining terminal comprises: a voice input unit for generating a voice frame by converting an inputted voice signal and outputting the voice frame; a voiced/unvoiced sound analysis unit for determining whether the voice frame inputted through the voice input unit corresponds to a voiced sound, an unvoiced sound, or background noise; a voice frame energy detection unit for extracting the average energy of voice frames which have been determined as a voiced sound by the voiced/unvoiced sound analysis unit; an interval energy detection unit for detecting the average energy of intervals including a plurality of voice frames which have been determined as voiced sounds; and an alcohol consumption determining unit for determining whether a person is drunk after consuming alcohol by extracting a difference value among the average energy of neighboring intervals which have been detected by the interval energy detection unit, thereby determining whether a person is drunk after consuming alcohol by analyzing the voice signal in a time domain.

Claims

1-10. -10. (canceled)

11. A method for determining whether a person is drunk comprising:

converting a voice signal received from said person into a plurality of voice frames;

extracting a first average energy for each of the voice frames;

dividing the plurality of voice frames into sections with a predetermined length;

calculating a second average energy for a plurality of voice frames included in each of the sections; and

determining whether said person is drunk by computing differences of the second average energy between neighboring sections.

12. The method of claim 11, wherein the converting of a received voice signal into a plurality of voice frames comprises:

extracting predetermined features from a voice frame among the plurality of voice frames, and

determining whether said voice frame is from a voiced sound, an unvoiced sound, or background noise.

13. The method of claim 12, wherein the predetermined features comprise periodic characteristics of harmonics, root mean square energy (RMSE), or zero-crossing count (ZC) of a low-band voice signal energy area.

14. The method of claim 12, wherein the determining whether said voice frame is from a voiced sound, an unvoiced sound, or background noise comprises using neural network.

15. The method of claim 12, wherein the extracting a first average energy for each of the voice frames comprises extracting a first average energy for each voice frame corresponding to the voiced sound.

16. The method of claim 11, wherein the sections, each of which comprises one or more voice frames, either overlap partially or do not overlap with each other.

17. The method of claim 11, wherein the determining whether said person is drunk by computing differences of the second average energy between neighboring sections comprises:

identifying a section and one or more neighboring sections thereof,

computing differences of the second average energy between the identified sections, and

determining whether said person is drunk according to the computed differences of the second average energy.

18. The method of claim 17, wherein the determining whether said person is drunk according to the computed differences of the second average energy comprises:

determining that said person is drunk when a difference in the second average energy between the neighboring sections is less than a predetermined threshold, and

determining that said person is not drunk when the difference is greater than the predetermined threshold.

19. The method of claim 17, wherein the identified sections comprise a section and one neighboring section, and the neighboring section either overlaps partially or does not overlap with said section.

30. A computer-readable recording medium having a computer program recorded thereon for performing the method of claim 11 to determine whether a person is drunk.

20. A terminal for determining whether a person is drunk comprising:

a voice input unit configured to convert a voice signal received from said person into voice frames and output the voice frames;

a voiced/unvoiced sound analysis unit configured to determine whether each of the voice frames corresponds to a voiced sound, an unvoiced sound, or background noise;

a voice frame energy detection unit configured to extract a first average energy of a voice frame that is determined as a voiced sound by the voiced/unvoiced sound analysis unit;

a section energy detection unit configured to calculate a second average energy for a section in which a plurality of voice frames determined as voiced sounds are included; and

an alcohol consumption determination unit configured to determine whether said person is drunk by computing differences of the second average energy between neighboring sections.

21. The terminal of claim 20, wherein the voiced/unvoiced sound analysis unit comprises:

a feature extraction unit extracting predetermined features from a voice frame among the plurality of voice frames, and

a recognition and determination unit determining whether said voice frame is from a voiced sound, an unvoiced sound, or background noise.

22. The terminal of claim 21, wherein the predetermined features comprise periodic characteristics of harmonics, root mean square energy (RMSE), or zero-crossing count (ZC) of a low-band voice signal energy area.

23. The terminal of claim 21, wherein the recognition and determination unit uses neural network.

24. The terminal of claim 20, wherein the section having one or more voice frames is defined by a predetermined length.

25. The terminal of claim 24, wherein a defined section either overlaps partially or does not overlap with another defined section.

26. The terminal of claim 20, wherein the alcohol consumption determination unit comprises a storage unit configured to prestore a threshold to determine whether said person is drunk and a difference calculation unit configured to compute differences of the second average energy between neighboring sections.

27. The terminal of claim 26, wherein the difference calculation unit computes differences of the second average energy between neighboring sections that partially overlap with each other or between neighboring sections that do not overlap with each other.

28. The terminal of claim 27, wherein the neighboring sections comprise a section selected by the difference calculation unit and one neighboring section thereof.

29. The terminal of claim 20, wherein the voice input unit receives the voice signal through a microphone provided therein or receives the voice signal from a remote site to generate the voice frame.