APPARATUS AND METHOD OF DIAGNOSING HEALTH BY USING VOICE

• US 20120220899A1
• Filed: 02/16/2012
• Published: 08/30/2012
• Est. Priority Date: 02/28/2011
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: KIM, Young-tae, SHIM, Hwan, OH, Hee-jin
• Granted Patent: US 9,055,861 B2
• Time in Patent Office: Days
• US Class Current: 600/586
• CPC Class Codes: A61B 5/4082 Diagnosing or monitoring movement diseases, e.g. Parkinson, Huntington or Tourette
• CPC Class Codes: A61B 5/4803 Speech analysis specially adapted for diagnostic purposes
• CPC Class Codes: A61B 5/7246 using correlation, e.g. template matching or determination of similarity
• CPC Class Codes: A61B 7/00 Instruments for auscultation
• CPC Class Codes: A61B 7/003 Detecting lung or respiration noise
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R

Abstract

A method and apparatus for diagnosing a user'"'"'s health state are provided. The apparatus includes including a voice detecting unit which detects and monitors a user'"'"'s voice; a voice analyzing unit which extracts a voice feature from a voice detected by the voice detecting unit, based on a health state to be diagnosed; a voice diagnosing unit which diagnoses a health state of the user by comparing the voice feature extracted by the voice analyzing unit with an abnormal state reference, and which monitors a change in the health state; and a diagnosis outputting unit which outputs information regarding the health state and a health state change diagnosed by the voice diagnosing unit.

Claims

1. A health diagnosing apparatus comprising:

a voice detecting unit which detects and monitors a user'"'"'s voice;

a voice analyzing unit which extracts a voice feature from a voice detected by the voice detecting unit, based on a health state to be diagnosed;

a voice diagnosing unit which diagnoses a health state of the user by comparing the voice feature extracted by the voice analyzing unit with an abnormal state reference, and which monitors a change in the health state; and

a diagnosis outputting unit which outputs information regarding the health state and the health state change diagnosed by the voice diagnosing unit.

2. The health diagnosing apparatus of claim 1, wherein the voice detecting unit monitors changes in a voice state of the user.

3. The health diagnosing apparatus of claim 1, wherein the voice detecting unit comprises:

an analog-to-digital converter; and

a voice detecting unit which detects a voice signal from a digital signal output by the analog-to-digital converter.

4. The health diagnosing apparatus of claim 1, wherein the voice analyzing unit comprises:

a health state selecting unit which selects a health state type based on the voice detected by the voice detecting unit;

a voice period selecting which selects a voice period for a diagnosis based on the health state type selected by the health state selecting unit;

a voice feature selecting unit which selects a voice feature according to the health state type selected by the health state selecting unit; and

a voice feature extracting unit which extracts for extracting the voice feature from the voice of the user based on the voice feature selected by the voice feature selecting unit.

5. The health diagnosing apparatus of claim 4, wherein the voice feature extracting unit comprises a normal voice database and an abnormal voice database.

6. The health diagnosing apparatus of claim 1, wherein the voice diagnosing unit comprises:

a comparing unit which compares the voice feature extracted by the voice analyzing unit with the abnormal state reference;

a monitoring unit which detects a change over time in a voice state of the user; and

a diagnosis processing unit which diagnosis a state of the user'"'"'s health according to a result of the comparison by the comparing unit, and the change detected by the monitoring unit.

7. The health diagnosing apparatus of claim 6, further comprising a storage unit which stores an output of the diagnosing processing unit and diagnosis time information.

8. The health diagnosing apparatus of claim 1, wherein the diagnosis outputting unit comprises:

a digital signal processor which converts information regarding the health state and a health state change diagnosed by the voice diagnosing unit into one or more audio or video signals;

a display unit which outputs a video signal converted by the digital signal processor; and

a speaker which outputs an audio signal converted by the digital signal processor.

9. A method of diagnosing a user'"'"'s health, the method comprising:

detecting and monitoring a voice of the user;

extracting a voice feature from the detected voice, based on a health state type to be diagnosed;

diagnosing a health state of the user by comparing the extracted voice feature with an abnormal state reference, and monitoring a change in the health state; and

outputting information regarding the health state and the health state change.

10. The method of claim 9, wherein the detecting of the voice comprises:

converting an input voice into a digital voice signal; and

detecting the voice of the user from the digital voice signal.

11. The method of claim 9, wherein the extracting of the voice feature comprises:

determining the health state type to be diagnosed;

selecting a voice period for a diagnosis according to the health state type;

selecting a voice feature corresponding to the health state type; and

extracting the selected voice feature from the selected voice period.

12. The method of claim 9, wherein the diagnosing of the health state comprises:

analyzing a change over time in the health state; and

outputting an advising message according to the change over time in the health state.

13. The method of claim 9, wherein the diagnosing of the health state and the monitoring of the health state change comprise:

comparing the extracted voice feature with the abnormal state reference;

detecting a change over time in a voice state; and

diagnosing states of the user'"'"'s health according to a result of the comparison, and the change in the voice state.

14. The method of claim 13, further comprising storing and updating a result of the diagnosing and diagnosis time information in a storage unit.

15. The method of claim 13, further comprising deriving references with respect to an abnormal voice and a normal voice.

16. The method of claim 15, wherein the deriving of the references comprises:

forming an abnormal voice database and a normal voice database;

extracting an abnormal voice feature and a normal voice feature from the abnormal voice database and the normal voice database, respectively;

performing comparison training with respect to the extracted abnormal voice feature and the extracted normal voice feature; and

deriving the abnormal state reference according to the comparison training

17. The method of claim 9, wherein the outputting comprises:

converting information regarding the health state and the health state change into one or more audio or video signals; and

outputting the one or more audio or video signals.

18. A computer-readable recording medium having recorded thereon a program for executing a method of diagnosing a user'"'"'s health, the method comprising:

detecting and monitoring a voice of the user;

extracting a voice feature from the detected voice, based on a health state type to be diagnosed;

diagnosing a health state of the user by comparing the extracted voice feature with an abnormal state reference, and monitoring a change in the health state; and

outputting information regarding the health state and a health state change.

19. A health diagnosing apparatus comprising:

a voice detector which detects a user'"'"'s voice;

a voice analyzer which extracts a voice feature from the voice detected by the voice detector;

a diagnoser which determines a health state of the user by comparing the voice feature extracted by the voice analyzer to an abnormal state reference and which monitors a change in the health state of the user; and

an output which outputs information regarding the health state of the user and a change in the health state of the user.

20. The health diagnosing apparatus of claim 19, wherein the voice feature extracted by the diagnoser is one or more of Jitter, Shimmer, a Harmonics-to-Noise ratio, and a Mel-Frequency Cepstral Coefficient.