| 摘要 |
A simply configured voice data selector and the like for producing a natural synthetic speech at high speed. When data representing a message template is supplied, a voice piece editor of the voice data selector searches a voice piece database for voice piece data on a voice piece whose sound matches a voice piece in the message template. Further, the voice piece editor predicts the cadence of the message template and selects, one at a time, a best match of each voice piece in the message template from the voice piece data that has been retrieved, according to an evaluation expression. For variables, the evaluation expression contains the result of primary regression of the frequency of the pitch component between the cadence prediction result and the voice piece data and the time difference in speech speeds. The voice piece data that is identified and the waveform data that is alternatively supplied from the acoustic processor because no voice piece data is identified are combined to generate data representing a synthetic speech.
|
