摘要 |
PURPOSE:To provide the storage voice rule synthesis device which generates no unnatural synthesized voice and is economical and efficient. CONSTITUTION:This voice rule synthesis device is equipped with a source sentence information storage means A stored with source sentence information S1, a source sentence information converting means 4 which converts the source information S1 in character code into phoneme information S2, a phoneme series information storage means B stored with the converted phoneme series information S2, a phoneme series information converting means 5 which converts the phoneme series information S2 into voice waveform information S3, a voice waveform information storage means c stored with the voice waveform information S3, a switching circuit D which selects and reads out all or some of the source sentence information S1, phoneme series information S2, and voice waveform information S3, a control part E which specifies and controls in which of the source sentence information storage part A, phoneme series information storage part B, and voice waveform information storage part C the source sentence information S1 is stored selectively according to the reference frequency of the request to voice out the source sentence information and also controls the switching circuit D, and a voice output means 6 outputs the voice waveform information S3. |