摘要 |
The purpose of this invention is to provide a speech generation device, a speech generation method, and a program whereby, from text data in Japanese, a musical phrase that carries the linguistic meaning of said Japanese is generated. This speech generation device is characterized by having the following: a sequence-code extraction means that extracts, from a sequence-code table that stores sequence codes corresponding to inputted text data, the sequence code corresponding to given inputted text data; a musical-note-code extraction means that extracts, from a musical-note-code table that stores musical-note codes corresponding to sequence codes, the musical-note codes corresponding to the aforementioned extracted sequence code; and a speech-data generation means that generates speech data from the extracted musical-note codes. |