パ-ソナルコンピュ-タを用いた音声情報処理システム

概要

論文の詳細を見る
Personal computers are more efficient than main frame computers in terms of portable and personal use, on-line processing, and conversational operation with a graphic display. To exemplify this, we present a new speech information processing system, which incorporates a speech analysis sub-system and a spoken word recogintion sub-system.The speech analysis sub-system can be operated easily and repeatedly, because it adopts a light-pen and the menu method. It can also play back any segment of sound from any part of the speech. By using this function, we can get new knowlege of the auditory response to speech sound.The spoken word recognition sub-system displays a running process of the system on the graphic display, and it makes it easier to understand the structure of speech recognition. The system is useful for demonstration of speech recognition and serves as a basis for developing practical systems, because it can recognize ten numbers, from zero to nine, uttered by males at a rate of almost 100%.
琉球大学工学部の論文