著者

米澤 朋子 鈴木 紀子 安部 伸治 間瀬 健二 小暮 潔

出版者

一般社団法人情報処理学会

雑誌

情報処理学会研究報告音楽情報科学(MUS)

巻号頁・発行日

vol.2008, no.50, pp.25-30, 2008-05-21

---

擬人的媒体等とのインタラクションに伴う音声表現に,連続的な声色表情付与を狙いとし,STRAIGHT を用いた音声モーフィングを同一歌唱者の表情無し歌声-表情付き間歌声間に適用した.合成音の声色の表情強度に関して一対比較を行った結果,第一刺激に比べて第二刺激の表情強度が増す場合と減る場合で,刺激間の差分の知覚が異なることがわかった.これに基づき声色の表情付け手法を議論する.This paper proposes and evaluate a method for synthesizing continuous expressions in vocal timbre by gradually changing spectral parameters based on STRAIGHT speech morphing algorithm among differently expressed singing voices. In order to synthesize natural change of various and continuous strengths of singing voice expressions, a singing voice without expression, "normal," is used as the base of morphing, and singing voices of the particular singer with three different expressions, "dark," "whispery" and "wet," are used as targets. The results of the paired comparison between different strengths of vocal timbre suggested that the decrease of the expression in vocal timbre is more perceived than the increase in some expressions.