ダイナミックアナログ音声合成装置

概要

論文の詳細を見る
This study shows our success of a new trial to accomplish the speech synthesizer which can speak continuously in real time through the simulator that assumes speech mechanism with the aid of an analog computer. As the analog computer hitherto, however, was very low in computing speed, it was difficult to make a real time operation in the speech frequency band. So we started, first, to develop an ultra high speed analog computer. The usable frequency band of this computer counts about 30 kHz. Through this computer it was possible to synthesize Japanese vowels, comparatively good, composing a vocal tract simulator. Next, on the synthesis of Japanese consonants, a dynamic control was exercised over the parameters corresponding to configurations, by the use of high speed analog multiplier newly developed. As a result, control noises are rejected and smooth control is possible till at last the Japanese syllables came all to synthesize. The control devices of the vocal tract simulator consists of a configuration control signal generator and a buzz and hiss generator which can set up articulatory parameters. And the out-put signals of two generaters control to the vocal tract simulator and then voice can be synthesized.
明治大学の論文