This study was conducted to use it as basic information to suggest the necessity and program for self-regulation that can properly maintain voice according to the degree of hearing loss. For this purpose, twenty-six men in their fifties were classified according to the degree of hearing loss, and multidimensional voice program (MDVP) (F<sub>0</sub>, jitter, shimmer, noise to harmonics ratio [NHR]), cepstrum (cepstral peak prominence [CPP], low/high spectral ratio [L/H ratio], mean CPP F<sub>0</sub>, cepstral/spectral index of dysphonia [CSID]) and vocal fatigue index (fatigue, physical, rest, total) were performed and analyzed. The more severe the degree of hearing loss, the higher the F<sub>0</sub> and CSID, and the lower the L/H ratio, the overall quality of the voice deteriorated. Normal-mild hearing loss group indicated a positive correlation between mean CPP F<sub>0</sub> and fatigue, rest and total, and negative correlation between CPP and rest. In severe-profound group, F<sub>0</sub> and rest showed a negative correlation, and shimmer and rest showed a positive correlation. Through the analysis of vowel and connected speech, it was found that the more severe the degree of hearing loss, the higher the F<sub>0</sub> in vowel phonation, the lower the L/H ratio and the higher the CSID, and the lower the overall sound quality. In addition, it was observed that that the more severe the hearing loss, the higher the fatigue level of the hearing-loss person who speaks due to severe voice misuse and abuse. Therefore, auditory rehabilitation patients with hearing loss optimal voice production, and voice rehabilitation programs that use proper voices should be supported.