Limits...
Study of a vocal feature selection method and vocal properties for discriminating four constitution types.

Kim KH, Ku B, Kang N, Kim YS, Jang JS, Kim JY - Evid Based Complement Alternat Med (2012)

Bottom Line: Further, we suggest a process to extract independent variables by eliminating explanatory variables and reducing their correlation and remove outlying data to enable reliable discriminant analysis.Moreover, the suitable division of data for analysis, according to the gender and age of subjects, is discussed.Finally, the vocal features are applied to a discriminant analysis to classify each constitution type.

View Article: PubMed Central - PubMed

Affiliation: Division of Constitutional Medicine Research, Korea Institute of Oriental Medicine, 461-24 Jeonmin-dong, Yuseong-gu, Daejeon 305-811, Republic of Korea.

ABSTRACT
The voice has been used to classify the four constitution types, and to recognize a subject's health condition by extracting meaningful physical quantities, in traditional Korean medicine. In this paper, we propose a method of selecting the reliable variables from various voice features, such as frequency derivative features, frequency band ratios, and intensity, from vowels and a sentence. Further, we suggest a process to extract independent variables by eliminating explanatory variables and reducing their correlation and remove outlying data to enable reliable discriminant analysis. Moreover, the suitable division of data for analysis, according to the gender and age of subjects, is discussed. Finally, the vocal features are applied to a discriminant analysis to classify each constitution type. This method of voice classification can be widely used in the u-Healthcare system of personalized medicine and for improving diagnostic accuracy.

No MeSH data available.


Related in: MedlinePlus

Vocal features of a sentence.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3313581&req=5

fig3: Vocal features of a sentence.

Mentions: The voice features are shown in Figures 2 and 3 for the five vowels and a sentence, respectively, in order not to lose any vocal information. The features of each vowel are T0, F0 (average pitch frequency), DTF0 (average difference of F0 over the time interval), F1–F4, BW1, BW2 (the formant frequencies and bandwidths of the 1st and 2nd terms) [16], F2/F1, F3/F1, F4/F1, F3/F2, F4/F2, F4/F3 (the ratios of the formant frequencies), JITA, JITT, PPQ, RAP (jitter, percentage of jitter, average variation, and average of pitch frequency), MFCC1-13 (the terms of mel-frequency cepstral coefficient (MFCC)), which are useful in the recognition of voice patterns [17–21], SHDB, SHIM, APQ (shimmer in dB, its variation, and the smooth variation of amplitude), the energy and the power of vowels, and the ratios of voice energies over fixed frequency bands, such as 60–120 Hz, 120–240 Hz, 240–480 Hz, 480–960 Hz, 960–1920 Hz, and 1920–3840 Hz.


Study of a vocal feature selection method and vocal properties for discriminating four constitution types.

Kim KH, Ku B, Kang N, Kim YS, Jang JS, Kim JY - Evid Based Complement Alternat Med (2012)

Vocal features of a sentence.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3313581&req=5

fig3: Vocal features of a sentence.
Mentions: The voice features are shown in Figures 2 and 3 for the five vowels and a sentence, respectively, in order not to lose any vocal information. The features of each vowel are T0, F0 (average pitch frequency), DTF0 (average difference of F0 over the time interval), F1–F4, BW1, BW2 (the formant frequencies and bandwidths of the 1st and 2nd terms) [16], F2/F1, F3/F1, F4/F1, F3/F2, F4/F2, F4/F3 (the ratios of the formant frequencies), JITA, JITT, PPQ, RAP (jitter, percentage of jitter, average variation, and average of pitch frequency), MFCC1-13 (the terms of mel-frequency cepstral coefficient (MFCC)), which are useful in the recognition of voice patterns [17–21], SHDB, SHIM, APQ (shimmer in dB, its variation, and the smooth variation of amplitude), the energy and the power of vowels, and the ratios of voice energies over fixed frequency bands, such as 60–120 Hz, 120–240 Hz, 240–480 Hz, 480–960 Hz, 960–1920 Hz, and 1920–3840 Hz.

Bottom Line: Further, we suggest a process to extract independent variables by eliminating explanatory variables and reducing their correlation and remove outlying data to enable reliable discriminant analysis.Moreover, the suitable division of data for analysis, according to the gender and age of subjects, is discussed.Finally, the vocal features are applied to a discriminant analysis to classify each constitution type.

View Article: PubMed Central - PubMed

Affiliation: Division of Constitutional Medicine Research, Korea Institute of Oriental Medicine, 461-24 Jeonmin-dong, Yuseong-gu, Daejeon 305-811, Republic of Korea.

ABSTRACT
The voice has been used to classify the four constitution types, and to recognize a subject's health condition by extracting meaningful physical quantities, in traditional Korean medicine. In this paper, we propose a method of selecting the reliable variables from various voice features, such as frequency derivative features, frequency band ratios, and intensity, from vowels and a sentence. Further, we suggest a process to extract independent variables by eliminating explanatory variables and reducing their correlation and remove outlying data to enable reliable discriminant analysis. Moreover, the suitable division of data for analysis, according to the gender and age of subjects, is discussed. Finally, the vocal features are applied to a discriminant analysis to classify each constitution type. This method of voice classification can be widely used in the u-Healthcare system of personalized medicine and for improving diagnostic accuracy.

No MeSH data available.


Related in: MedlinePlus