摘要 |
PURPOSE:To structure a multimedia network which assists conversion into a voice at respective speech synthesis parts on local computers by equipping one computer with a language processing part and other computers with the speech synthesis parts, and performing decentralized processing. CONSTITUTION:Only one computer, i.e., a remote machine 2 is equipped with the language processing part 4 and other computers, i.e., local machines 3 are equipped with the speech synthesis parts 5. Respective users have only the speech synthesis parts 5 on their local machines 3 at hand, and the language processing part 4 operates on only one machine, i.e., the remote machine 2 on the network 1, so that the decentralized processing is performed. Consequently, a KANJI(Chinese character)-KANA(Japanese syllabary) mixed sentence is sent to the language processing part 4 on the remote machine 2, and the speech synthesis parts 5 on the local machines 3 receive the analytic output results and synthesize speeches, which can be outputted. Therefore, each machine need not be equipped with the language processing part, and a speech synthesis system which is efficient including cost is obtained. |