1A2-C12 人工喉頭を用いたSinger Robotの研究 : 母音フォルマントのリアルタイム最適調整(アミューズメント・エンタテイナーロボット)

概要

論文の詳細を見る
Now, voices of almost all robots are made by speech synthesizer. They are sounds produced by electronic speakers. This is a study on Singer Robot using artificial larynx. Now it can generate Japanese 5 vowels [a], [i], [u], [e], and [o]. But it is very difficult to generate clear vowels because the frequencies of formants are not easy to be controlled by changing the cross sectional area of the vocal tract. Singer Robot is expected to produce vowels with correct formants in order to sing in distinct pronunciations. We proposed a new optimal adjustment method of vowel formant. In this method, an optimal area distribution of vocal tract can be searched for the targeted vector of the first formant and the second formant of clear vowel. In this paper, we introduce a system to adjust formants optimally, which analyzes formants in real-time and feeds back them to Singer Robot.
2007-05-11