The influence of coarticulatory and phonemic relations on individual compensatory formant production.

Previous auditory perturbation studies have shown that speakers are able to simultaneously use multiple compensatory strategies to produce a certain acoustic target. In the case of formant perturbation, these findings were obtained examining the compensatory production for low vowels /ɛ/ and /æ/. This raises some controversy as more recent research suggests that the contribution of the somatosensory feedback to the production of vowels might differ across phonemes. In particular, the compensatory magnitude to auditory perturbations is expected to be weaker for high vowels compared to low vowels since the former are characterized by larger linguopalatal contact. To investigate this hypothesis, this paper conducted a bidirectional auditory perturbation study in which F2 of the high central vowel /ɨ/ was perturbed in opposing directions depending on the preceding consonant (alveolar vs velar). The consonants were chosen such that speakers' usual coarticulatory patterns were either compatible or incompatible with the required compensatory strategy. The results demonstrate that speakers were able to compensate for applied perturbations even if speakers' compensatory movements resulted in unusual coarticulatory configurations. However, the results also suggest that individual compensatory patterns were influenced by additional perceptual factors attributable to the phonemic space surrounding the target vowel /ɨ/.

The coarticulation/invariance scale: mutual information as a measure of coarticulation resistance, motor synergy, and articulatory invariance.

Coarticulation and invariance are two topics at the center of theorizing about speech production and speech perception. In this paper, a quantitative scale is proposed that places coarticulation and invariance at the two ends of the scale. This scale is based on physical information flow in the articulatory signal, and uses Information Theory, especially the concept of mutual information, to quantify these central concepts of speech research. Mutual Information measures the amount of physical information shared across phonological units. In the proposed quantitative scale, coarticulation corresponds to greater and invariance to lesser information sharing. The measurement scale is tested by data from three languages: German, Catalan, and English. The relation between the proposed scale and several existing theories of coarticulation is discussed, and implications for existing theories of speech production and perception are presented.

Motor equivalent strategies in the production of German /∫/ under perturbation.

The German sibilant /∫/ is produced with a constriction in the postalveolar region and often with protruded lips. By covarying horizontal lip and tongue position speakers can keep a similar acoustic output even if the articulation varies. This study investigates whether during two weeks of adaptation to an artificial palate speakers covary these two articulatory parameters, whether tactile landmarks have an influence on the covariation and to what extent speakers can foresee the acoustic result of the covariation without auditory feedback. Six German speakers were recorded with EMA. Four of them showed a covariation of lip and tongue, which is consistent with the motor equivalence hypothesis. The acoustic output, however, does not stay entirely constant but varies with the tongue position. The role of tactile landmarks is negligible. To a certain extent, speakers are able to adapt even without auditory feedback.

Adaptation strategies in perturbed /s/.

The purpose of this work is to investigate the role of three articulatory parameters (tongue position, jaw position and tongue grooving) in the production of /s/. Six normal speakers' speech was perturbed by a palatal prosthesis. The fricative was recorded acoustically and through electromagnetic articulography in four conditions: (1) unperturbed, (2) perturbed with auditory feedback masked, (3) perturbed with auditory feedback available and (4) perturbed after a 2-week adaptation period. At the end of the adaptation, speakers produced more high-frequency noise while either having a higher jaw position or more grooving of the tongue or both. We discuss the potential clinical implications of the results with regard to the role of jaw height and tongue grooving in the treatment of impaired /s/.

Human vocal tract analysis by in vivo 3D MRI during phonation: a complete system for imaging, quantitative modeling, and speech synthesis.

We present a complete system for image-based 3D vocal tract analysis ranging from MR image acquisition during phonation, semi-automatic image processing, quantitative modeling including model-based speech synthesis, to quantitative model evaluation by comparison between recorded and synthesized phoneme sounds. For this purpose, six professionally trained speakers, age 22-34y, were examined using a standardized MRI protocol (1.5 T, T1w FLASH, ST 4mm, 23 slices, acq. time 21s). The volunteers performed a prolonged (> or = 21s) emission of sounds of the German phonemic inventory. Simultaneous audio tape recording was obtained to control correct utterance. Scans were made in axial, coronal, and sagittal planes each. Computer-aided quantitative 3D evaluation included (i) automated registration of the phoneme-specific data acquired in different slice orientations, (ii) semi-automated segmentation of oropharyngeal structures, (iii) computation of a curvilinear vocal tract midline in 3D by nonlinear PCA, (iv) computation of cross-sectional areas of the vocal tract perpendicular to this midline. For the vowels /a/,/e/,/i/,/o/,/ø/,/u/,/y/, the extracted area functions were used to synthesize phoneme sounds based on an articulatory-acoustic model. For quantitative analysis, recorded and synthesized phonemes were compared, where area functions extracted from 2D midsagittal slices were used as a reference. All vowels could be identified correctly based on the synthesized phoneme sounds. The comparison between synthesized and recorded vowel phonemes revealed that the quality of phoneme sound synthesis was improved for phonemes /a/, /o/, and /y/, if 3D instead of 2D data were used, as measured by the average relative frequency shift between recorded and synthesized vowel formants (p < 0.05, one-sided Wilcoxon rank sum test). In summary, the combination of fast MRI followed by subsequent 3D segmentation and analysis is a novel approach to examine human phonation in vivo. It unveils functional anatomical findings that may be essential for realistic modelling of the human vocal tract during speech production.

Speaker-specific kinematic properties of alveolar reductions in English and German.

A simultaneous EPG/EMA study of tongue gestures of five speakers was conducted to investigate the kinematic events accompanying alveolar stop reductions in the context of a velar plosive /k/ and in the context of a laryngeal fricative /h/ in two languages, English and German. No systematic language differences could be detected. Alveolar productions before a following /h/ showed only a marginal weakening of the formation of complete occlusion, while alveolar productions before a following /k/ showed a wide range of reductions, including instances of a complete deletion of the alveolar gesture. The extension of movement reduction varied between and within subjects. Importantly, while speakers were consistent with themselves, they employed different articulatory patterns with respect to the timing relationship between movement initiation, overall movement duration, peak velocity as well as closure duration. An attempt is made to relate the observed movement patterns to the dynamic factors of the speech mechanism.

The influence of oral cavity tumour treatment on the voice quality and on fundamental frequency.

This study aimed to discover whether the surgical treatment of oral cavity tumours only affects on articulation or whether it also leads to a change in voice quality and fundamental frequency. Twelve participants were examined pre- and postoperatively for mean speaking fundamental frequency, standard deviation of the mean fundamental frequency, harmonic-to-noise-ratio and intrinsic pitch. All the parameters showed a substantial postoperative change in some patients.