摘要 |
PROBLEM TO BE SOLVED: To provide a speech synthesizer permitting to give spoken guidance with accents as proper as possible even for the information on reading KANA (Japanese character) such as those for place names, intersection names. SOLUTION: This is a speech synthesizer provided with a storage medium in which information on reading KANA transcribing the reading of the names of places, etc., with KANA are stored, a reading KANA information extracting means 4 for extracting the above-mentioned reading KANA information from this storage medium, a speech synthesizing means 7 for converting the extracted KANA into speech, and a language analyzing means 8 to give accents to the reading KANA extracted by the reading KANA information extracting means 4. In such a case, the language analyzing means 8 is to give the accent to the above-mentioned reading KANA information based on the result of analysis of compound word relation in the above-mentioned reading KANA information. |