摘要 |
<p>PROBLEM TO BE SOLVED: To generate a synthesized voice at high speed and easy to catch according to a sentence. SOLUTION: When waveform data ((1), (2), (3), etc.), corresponding to each character included in a sentence are composited, a part of waveform data (for example, the second half of (4)) corresponding to preceding characters and a part of voice waveform corresponding to succeeding character (for example, the first half of (5)) are overlapped with each other.</p> |