| 摘要 |
PROBLEM TO BE SOLVED: To provide a read-aloud text data selecting apparatus for speech synthesis that can select as read-aloud text data only such text data as contains lacking speech data units.SOLUTION: A read-aloud text data selecting apparatus 1 for speech synthesis comprises speech data unit target number setting means 30 that references a speech data unit dictionary, counts speech data units already contained in read-aloud text data and sets the target number of lacking speech data units desired to be contained in the read-aloud text data, and read-aloud text data selecting means 40 that references the speech data unit dictionary and selects text data out of read-aloud text data candidates so that as many speech data units as the target number can be contained in the read-aloud text data. |
