Voice conversion is a technique to change speaker individuality
i.e., speech uttered by a speaker is changed to sound as if another speaker had uttered it. In this paper, we propose a voice conversion algorithm that uses speech segments as conversion units. Input speech is decomposed into speech segments by a speech recognition module and the segments are replaced by speech segments uttered by another speaker. This algorithm makes it possible to convert not only the static characteristics but also speaker individuality contained in phoneme segments. The proposed voice conversion algorithm was performed between two male speakers. Spectrum distortion between target speech and the converted speech was reduced to one-third the natural spectrum difference between the two speakers. A litening experiment showed that, in terms of speaker identification accuracy, the speech converted by segment-sized units gave a score 20% higher than the speech converted frame-by-frame. © 1992, Acoustical Society of Japan. All rights reserved.