Articulatory synthesis generates speech sounds by simulating the physical phenomena involved in speech production. The accuracy of the physical modelling is expected to affect the naturalness of the synthesis: the more realistic the description is, the greater the naturalness is expected to be. In this work, the accuracy of acoustic wave propagation in the vocal tract was evaluated with two perceptual experiments. Sustained vowels generated using a one-dimensional acoustic model, a three-dimensional acoustic model and an artificial bandwidth extension algorithm (without a physical basis) were compared. Since the difference between the acoustic methods tested affects mainly the frequencies above 4 kHz, we ensured that the low frequency part of the stimuli, up to 4 kHz, was similar. Thus, the participants' responses were based only on the differences at high frequency. The first experiment was a pair comparison, in which the participants had to select the more natural sounding stimuli. In the second experiment, the participants had to rate the naturalness of the stimuli on a linear scale. The results confirmed that a more accurate physical modeling leads to greater naturalness. However, this was limited to the phonemes /o/ and /u/, for which transverse resonances in the anterior vocal tract may play an important role that only a 3D acoustic simulation can accurately represent. It was also found that male stimuli were perceived as significantly more natural than female ones. However, voice quality did not affect naturalness.

abstract: Speech is generated by articulators acting on a phonatory source. Identification of this phonatory source and articulatory geometry are individually challenging and ill-posed problems, called speech separation and articulatory inversion, respectively. There exists a trade-off between decomposition and recovered articulatory geometry due to multiple possible mappings between an articulatory configuration and the speech produced. However, if measurements are obtained only from a microphone sensor, they lack any invasive insight and add additional challenge to an already difficult problem. A joint non-invasive estimation strategy that couples articulatory and phonatory knowledge would lead to better articulatory speech synthesis. In this thesis, a joint estimation strategy for speech separation and articulatory geometry recovery is studied. Unlike previous periodic/aperiodic decomposition methods that use stationary speech models within a frame, the proposed model presents a non-stationary speech decomposition method. A parametric glottal source model and an articulatory vocal tract response are represented in a dynamic state space formulation. The unknown parameters of the speech generation components are estimated using sequential Monte Carlo methods under some specific assumptions. The proposed approach is compared with other glottal inverse filtering methods, including iterative adaptive inverse filtering, state-space inverse filtering, and the quasi-closed phase method. / Dissertation/Thesis / Masters Thesis Electrical Engineering 2018

Articulatory synthesis is based on modeling various physical phenomena of speech production, including sound radiation from the mouth. With regard to sound radiation, the most common approach is to approximate it in terms of a simple spherical source of strength equal to the mouth volume velocity. However, because this approximation is only valid at very low frequencies and does not account for the diffraction by the head and the torso, we simulated two alternative radiation characteristics that are potentially more realistic: the radiation from a vibrating piston in a spherical baffle, and the radiation from the mouth of a detailed model of the human head and torso. Using the articulatory speech synthesizer VocalTractLab, a corpus of 10 sentences was synthesized with the different radiation characteristics combined with three different phonation types. The synthesized sentences were acoustically compared with natural recordings of the same sentences in terms of their long-term average spectra (LTAS), and evaluated in terms of their naturalness and intelligibility. The intelligibility was not affected by the type of radiation characteristic. However, it was found that the more similar their LTAS was to real speech, the more natural the synthetic sentences were perceived to be. Hence, the naturalness was not directly determined by the realism of the radiation characteristic, but by the combined spectral effect of the radiation characteristic and the voice source. While the more realistic radiation models do not per se improve synthesis quality, they provide new insights in the study of speech production and articulatory synthesis.