Voice　conversion　is　a　technique　used　to　transform　one　speaker＇s　voice　into　another　speaker＇s　voice.　This　paper　describes　a　study　on　voice　conversion　using　genetic　algorithm　(GA)　to　train　the　hidden　layer　of　RBF　neural　network,　which　is　expected　to　help　improve　the　preference　of　converted　speech　for　the　target　speaker＇s　characteristics.　Six　mono-vowel　phonemes　in　Mandarin　speech　were　used　for　conversion　experiments　which　were　performed　on　neural　networks　respectively　by　GA-based　and　K-means　methods.　Subjective　and　objective　evaluations　were　conducted　on　the　performances　of　converted　speech.　The　conversion　results　show　that　in　spite　of　not　too　much　improvement　in　perceptual　distance,　the　RBF　network　by　genetic　algorithm　instead　of　by　K-means　method　has　the　ability　of　global　optimization　with　an　evident　decrease　in　the　spectral　distance　between　the　converted　speech　and　the　target　speech.