OBJECTIVES Phonation is influenced by hearing as a feedback mechanism. The purpose of the present study was to compare selected acoustic parameters in children using cochlear implants (CIs), those using hearing aids (HA), and their normal-hearing (NH) peers. METHODS The participants were 15 children using CI (mean age: 72 months), 15 children using HA (mean age: 74 months), and 15 NH children (mean age: 77 months). The vowel /a/ was produced to measure perturbation and mean fundamental frequency. The six Persian vowels in /CbVCd/ were obtained to extract vowel duration. Data were analyzed by one-way analysis of variance. RESULTS Results revealed a statistically significant difference between the NH group and the HA group regarding fundamental frequency (F2,51 = 3.443, P < 0.05), jitter local (F2,51 = 1.629, P < 0.05), jitter local absolute (F2,51 = 6.519, P < 0.001), jitter rap (F2,51 = 7.151, P < 0.001), jitter ppq5 (F2,51 = 5.894, P < 0.001), shimmer local (%) (F2,51 = 8.070, P < 0.001), shimmer local (dB) (F2,51 = 3.884, P < 0.05), shimmer apq3 (F2,51 = 4.926, P < 0.05), shimmer apq5 (F2,51 = 8.442, P < 0.001), and harmonic-to-noise ratio (F2,51 = 4.117, P < 0.001). The mean values of the duration of all six vowels were significantly greater in children with CI and HA than in NH children (P < 0.001). CONCLUSION It seems that after 8 months of using CI, auditory control of voice production would be enabled. Furthermore, children with hearing impairment potentially regard vowel sound duration as a distinguishing feature, whereas in NH speakers, the duration has the least effect in vowel identification.