| 摘要 |
PROBLEM TO BE SOLVED: To provide a speech synthesis device capable of sufficient speech synthesis processing or mixed excitation even under a limited environment of computational performance, and to provide a speech synthesis method and a speech synthesis program.SOLUTION: A speech synthesis device 100 for synthesizing a speech waveform on the basis of inputted time series sound source control information and spectrum characteristic information, comprises: a subband-divided sound source generation section 120 for generating a subband-divided sound source waveform vector corresponding to the inputted sound source control information on the basis of subband-divided sound source waveform vectors that have been accumulated through dividing a sound source waveform into a plurality of frequency bands; a subband power adjustment section 130 for performing an amplitude adjustment according to the inputted spectrum characteristic information, on the generated subband-divided sound source waveform vector for each subband; and a subband synthesis section 140 for synthesizing the amplitude-adjusted subband-divided sound source waveform vector into a single speech waveform. |
