Natural language processing in a Japanese text-to-speech system forwritten-style texts

摘要

This paper introduces a Japanese text-to-speech (TTS) system,AUDIOTEX, that modifies written-style expressions for listening andgenerates prosodic information (Kanji reading, accent, and pause).Conventional TTS systems have several problems: most texts are writtento be read with the human eye, thus they are inadequate for listening;and Kanji reading errors and accentuation errors reduce intelligibility.To solve these problems, we have developed a rewriting process forlistening based on flexible tree conversion rules and a highly accuratemorphological analysis which retains word candidates with an errorcorrection environment