VOICE PROCESSING DEVICE, VOICE PROCESSING METHOD, AND VOICE PROCESSING PROGRAM

摘要

PROBLEM TO BE SOLVED: To provide a voice processing device not requiring an acoustic model.SOLUTION: A voice processing device comprises: an acquisition part for acquiring input voice; and an estimation part for estimating at least one vowel section length on the basis of a first transition point and a second transition point of an amount of change of sound feature quantity included in the input voice. Further, the voice processing device comprises a control part for controlling speed of the input voice on the basis of the vowel section length.