To provide a voice synthesis device in which interesting articles are selected in order of degree of interest from an article group, with conditions such as total voice time length, speaking speed and the number of articles which must be reproduced.
The voice synthesis device for generating synthesis voice from the selected article, comprises: a parameter setting means for setting the speaking speed of desired synthesis voice, speaker classification and the time length; a keyword setting means for setting a keyword and a degree of interest; an interest degree determination means for determining an interest degree, based on the keyword and the degree of interest which are set; a summary level determination means for determining a summary level of a text based on the determined interest degree; a text summary means for summarizing the text, based on the summary level; and a text voice synthesis means for synthesizing voice, based on the speaking speed and the classification which are set by the parameter setting means.
JPH09230892 | TEXT-SPEECH CONVERSION DEVICE |
JP2002189489 | SPEECH SYNTHESIZER |
JPH0667697 | SPEECH SYNTHESIZING DEVICE |
MIZUNO HIDEYUKI
JP2006126548A | 2006-05-18 | |||
JP2001282815A | 2001-10-12 | |||
JPH0863188A | 1996-03-08 | |||
JPH05232982A | 1993-09-10 | |||
JPH09190437A | 1997-07-22 | |||
JP2002108601A | 2002-04-12 |
Taku Kusano
Yukio Nakamura