音声解析装置、信号解析装置、音声解析システムおよびプログラム

摘要

PROBLEM TO BE SOLVED: To provide a voice analysis device capable of discriminating spoken voice of a wearer and spoken voice of others when the voice acquired by voice acquisition means is a mixture of the spoken voice of the wearer and the spoken voice of the others.SOLUTION: A terminal device 10 includes: voice signal acquisition means for acquiring voice signals generated by a voice acquisition means group composed of a set of adjacently arranged microphones arranged in plural separated by different distances from a vocalizing portion of a wearer; separation means for separating mixed voice signals for every speaker when the voice signals generated by the microphones are mixed voice signals of a plurality of speakers; and non-self discrimination means for discriminating whether or not the voice signals separated by the separation means are voice signals by the spoken voice of the wearer or the voice signals of ones other than the wearer.