^Roads 1996, p. 153, Figure 4.23: Overview of spectrum modeling synthesis.
^Serra & Smith 1990, p. 12. "It describes a technique called spectral modeling synthesis [SMS], that models time-varying spectra as (1) a collection of sinusoids controlled through time by piecewise linear amplitude and frequency envelopes (the deterministic part), and (2) a time-varying filtered noise component (the stochastic part). The analysis procedure first extracts the sinusoidal trajectories by tracking peaks in a sequence of short-time Fourier transforms. These peaks are then removed by spectral subtraction. The remaining “noise floor” is then modeled as white noise through a time-varying filter. A piecewise linear approximation to the upper spectral envelope of the noise is computed each successive spectrum, and the stochastic part is synthesized by mean of the overlap-add technique."
^McAulay & Quatieri 1988, p. 161, Fig. 8. "This block diagram of the sinusoidal analysis/synthesis system illustrates the major functions subsumed within the system. Neither voicing decisions nor residual waveforms are required for speech synthesis."
_in_Japanese.jpg)
_in_Japanese.jpg)
_in_Japanese.jpg)
.svg)
