摘要 |
PROBLEM TO BE SOLVED: To suppress an increase in processing time as much as possible and to prevent a voice output from becoming unnatural or indistinct as much as possible. SOLUTION: A phoneme sequence output part 2 decomposes input text information into phoneme symbols of phoneme sequences registered in a voice database and combines those registered phoneme sequences together to generate converted phoneme sequences. When an absent part of the input text information is included in all the converted phoneme sequences, an absent part count means 3 counts absent parts of the respective converted phoneme sequences to exclude converted phoneme sequences except the one having the smallest count value. |