음성장애 평가에서 켑스트럼 분석의 임상적 유용성 = Clinical Usefulness of Cepstral Analysis in Dysphonia Evaluation

다국어 초록 (Multilingual Abstract)

Background and Objectives With the present methods for assessing speech, there are procedures that make effective diagnostics possible for voice disorders. One such procedure is cepstrum. Spectrum is produced by Fourier transformation of sound waves, and if Fourier transformation is performed again about the spectrum, cepstrum is produced. This study compared cepstrum to jitter, shimmer, and noise-to-harmonic ratio (NHR) for assessment of its usefulness. Cepstrum was measured by Cepstral Peak Prominence (CPP) calculated by the Hillenbrand method and Cepstral Mean Values (CMV) calculated using the Computerized Speech Laboratory software.
Subjects and Method We included in the study 30 patients with vocal nodule and unilateral vocal cord palsy who were diagnosed in Kangbuk Samsung Hospital between March 2010 and May 2011, and 30 normal controls. Phonation of sustained vowel /a/ sample and running speech was subjected to acoustic analysis using CMV and CPP. Then we compared the correlation of cepstrum with other acoustic methods.
Results The measured values of CPP-a were 14.16, 17.25, 20.00 and the age adjusted CPP-s values were 11.21, 12.85, 15.00 for vocal cord palsy, vocal nodule and normal group, respectively. There was significant correlation with perceptions of dysphonia (p＜0.001), but in CMV-a, CMV-s, there was no significant correlation. When jitter, shimmer, and NHR were compared with cepstrum, the result showed negative correlation among the three groups but CPP values showed significant difference (p＜0.001).
Conclusion In the assessment of voice disorders, cepstrum may be used as a reliable method for comparing other complementary analysis tools. For the acoustic analysis of voice by cepstrum, however, CPP has more reliable correlations with dysphonia than CMV.

참고문헌 (Reference)

• 1 Horii Y, "Vocal shimmer in sustained phonation" 23 (23): 202-209, 1980
• 2 Heman-Ackah YD, "The relationship between cepstral peak prominence and selected parameters of dysphonia" 16 (16): 20-27, 2002
• 3 Kim KM, "Significance of acoustic parameters obtained from sound spectrogram in determining the phonatory function of the patients with vocal nodule" 32 (32): 860-866, 1989
• 4 Heman-Ackah YD, "Reliability of calculating the cepstral peak without linear regression analysis" 18 (18): 203-208, 2004
• 5 Titze IR, "Principles of voice production. 2nd ed" National Center for Voice and Speech 2000
• 6 Lieberman P, "Pertubations in vocal pitch" 33 : 334-342, 1963
• 7 Zieger K, "Cepstrum analysis in voice disorders" 47 (47): 210-217, 1995
• 8 Heman-Ackah YD, "Cepstral peak prominence: a more reliable measure of dysphonia" 112 (112): 324-333, 2003
• 9 Balasubramanium RK, "Cepstral analysis of voice in unilateral adductor vocal fold palsy" 25 (25): 326-329, 2011
• 10 Radish Kumar B, "Cepstral analysis of voice in persons with vocal nodules" 24 (24): 651-653, 2010
• 11 Kerst LG, "Amplitude cross-scetion representation with the sound spectrograph" 20 : 796-801, 1948
• 12 Hillenbrand J, "Acoustic correlates of breathy vocal quality: dysphonic voices and continuous speech" 39 (39): 311-321, 1996
• 13 Hillenbrand J, "Acoustic correlates of breathy vocal quality" 37 (37): 769-778, 1994
• 14 Hong KH, "Acoustic and laryngeal characteristics of Korean traditional singers" 39 (39): 1622-1632, 1996
• 15 Hong KH, "A study for acoustic parameters in unilateral vocal cord palsy" 35 (35): 334-345, 1992
• 16 Hillenbrand J, "A methodological study of perturbation and additive noise in synthetically generated voice signals" 30 (30): 448-461, 1987