Both the HYBRID method described here and the PSOLA method have been implemented in our laboratory as part of an experimental diphone speech synthesizer for multiple voices, including particularly, women and children. Informal listening comparisons of the two methods for pitch control led us to the subjective impression that the HYBRID method produced slightly, but noticeably, less distortion, especially when dealing with high-pitched speech. Since this subjective impression was consistent with reports that RELP methods outperformed PSOLA under at least some conditions, we sought to verify the impression with formal listening tests designed to compare the segmental intelligibility of speech processed by the two methods.