Mixed source model and its adapted vocal tract filter estimate for voice transformation and synthesis

A uniform phase representation for the harmonic model in speech synthesis applications

“…On the contrary, the impact of the glottal source on the vocal tract feature is far from negligible, which is not convenient. However, a robust separation of the vocal tract filter and the glottal source is far from straightforward [31,[41][42][43][44]. Thus, in this work, we chose to favor again robustness and simplicity, in order to focus, beforehand, on the phase features.…”

“…This is done through the following steps: first, aHM analysis [16] is performed to obtain the instantaneous phase from the waveform; then, the minimum-phase term is subtracted from the measured phases, and the local Phase Distortion (PD) [25] is calculated; finally, the short-time mean and standard deviation of the PD are computed in the neighborhood of each frame, the former being highly correlated to the maximum-phase component, and the latter to the degree of noisiness. Among the advantages of this novel approach, we can mention the following: (i) it is valid to analyze signals exhibiting harmonic and noise components that overlap both in time and in frequency and thus, avoiding binary voiced/unvoiced decisions which are error-prone and result in annoying artifacts, especially in synthesis [30,31]. (ii) Since it helps avoiding an explicit separation between harmonics and noise, it provides a solid and uniform framework for speech manipulation thus, avoiding artifacts near the voicing boundaries [21].…”

The role of glottal source parameters for high-quality transformation of perceptual age

“…Recent generations of speech synthesizers include an explicit representation of the glottal source (e.g., LF model [24,2]) in the speech synthesizer, which substantially improves the control of the voice quality during speech synthesis. In particular, the SVLN speech synthesizer (Separation of the Vocaltract with the Liljencrants-fant model plus Noise) [2] allows the intuitive control of the glottal source (and thus, voice quality) during speech synthesis [25], with a limited number of parameters. This section summarizes the main principles of the SVLN speech synthesizer.…”

“…The estimation of the SVLN parameters is described in details in [2]. Additionally, glottal closure instants (GCI) are estimated [27], and used to measure the period-to-period regularity of the glottal pulse :…”

“…In a time where speech synthesizers can perform high-quality transformation of a speech signal (e.g., STRAIGHT [1], SVLN [2]), the intuitive control of the voice transformation is highly desirable for extending the expressive repertoire of a voice [3] (typically, to create avatars in video-games and movies). To do so, statistical methods are used to model the speech parameters in order to control the voice identity [4], emotion [5], and "voice quality" [6,7] of a speech recording.…”