DOCUMENT VOCALIZING DEVICE

摘要

PURPOSE: To enable natural vocalization matching contexts and scenes at all times by A extracting a number character string which is present before and behind a special word detected from a Japanese analytic result and deciding a reading based upon a stored rule corresponding to the special word. CONSTITUTION: A special word detecting means 17 detects the special word stored in a storage means 12 according to the Japanese analytic result. An extracting means 8 extracts the number character string which is present nearby before and behind the detected special word. A 2nd storage means 21 stores data on a list of rules determined corresponding to plural special words stored in a 1st storage means 12. A voice data generating means 8 decides readings to the extracted number character string on the basis of the rule stored in the 2nd storage means 21 corresponding to the detected special word to generate voice data from the number character string. A control part 3 as a registration means registers the special word in the 1st storage means 12 and also registers its rule in the 2nd storage means 21.