The impact of congenital visual deprivation on speech production in adults was examined in an ultrasound study of compensation strategies for lip-tube perturbation. Acoustic and articulatory analyses of the rounded vowel /u/ produced by 12 congenitally blind adult French speakers and 11 sighted adult French speakers were conducted under two conditions: normal and perturbed (with a 25-mm diameter tube inserted between the lips). Vowels were produced with auditory feedback and without auditory feedback (masked noise) to evaluate the extent to which both groups relied on this type of feedback to control speech movements. The acoustic analyses revealed that all participants mainly altered F2 and F0 and, to a lesser extent, F1 in the perturbed condition - only when auditory feedback was available. There were group differences in the articulatory strategies recruited to compensate; while all speakers moved their tongues more backward in the perturbed condition, blind speakers modified tongue-shape parameters to a greater extent than sighted speakers.

Studies with congenitally blind speakers show that visual deprivation yields increased auditory discrimination abilities as well as reduced amplitude of labial movements involved in vowel production, compared with sighted speakers. To further investigate the importance of auditory and visual feedback in speech, a study of auditory perturbation of rounded vowels was conducted in congenitally blind and sighted French speakers. Acoustic and articulatory (electromagnetic articulography) recordings from ten congenitally blind speakers and ten sighted speakers were obtained during the production of the French rounded vowel /ø/. All participants were first asked to produce the vowel repeatedly in a normal condition, i.e., with regular auditory feedback. In the perturbed condition, participants received, in real-time through headsets, an altered version of their speech, in which F2 was gradually increased up to 500 Hz. To compensate for this perturbation, speakers had to enhance lip protrusion and/or tongue retraction. These adaptive maneuvers should have been concurrent with auditory perception abilities. Preliminary results show that congenitally blind speakers gave greater weight to auditory perception than their sighted peers, while compensating differently for the perturbations. These findings support the hypothesis that vision plays a significant role in the implementation of phonological targets.

Contrastive focus serves to emphasize the importance of a semantic unit in the language string. Children with autistic spectrum disorder (ASD) appear to show difficulties in producing this prosodic marker. This study aimed to identify acoustic correlates related to contrastive focus in children with ASD. Nine francophone children with ASD and nine francophone typically developing (TYP) children produced simple four-word sentences (for example, «C’est une chaise.»: «it is a chair.») in a neutral condition and then in a contrastive focus condition. Ninety-six speech productions were recorded using a system that synchronized acoustic signals with lingual and labial movements. Maximum pitch, mean pitch, and pitch range, as well as maximum and mean sound intensity and duration were investigated. Values for pitch range, maximum and mean sound intensity, and duration were greater in the focus condition than in the neutral condition. Moreover, the differences were significantly greater in TYP children than in ASD children, who did not have increased speech values when switching to the focus mode. This suggests that pitch range, and intensity and duration of sound correlate most with contrastive focus marking in both groups. Yet, it appears that ASD children show less contrastive focus marking than TYP children.

Speech production entails appropriately timed contractions of many muscles. Steinert myotonic dystrophy, a neurodegenerative disease that causes muscle weakness and difficulties in muscle relaxation after muscle contraction, frequently affects orofacial articulatory dynamics leading to decreased speech intelligibility. We aimed to investigate the articulatory and acoustic characteristics of cardinal vowels produced by children with Steinert disease. We recruited fourteen 6- to 14-year-old French-speaking children diagnosed with Steinert disease and 14 aged-matched typically developing children. They were asked to produce repetitions of the vowels /i a u/ in consonant-vowel (CV) contexts. A synchronized ultrasound, Optotrak motion tracking system, and audio recording system was used to track lip and jaw displacement as well as tongue shape and position. Duration and formant values were also extracted. The Euclidean distance between vowels, in the formant space, was reduced in children with Steinert disease compared to the control children. Different patterns of articulatory contrasts were observed among the children, with some of them using more tongue contrasts than lip contrasts. Intelligibility tests conducted with adult listeners on a subset of the data show that some patterns are related to higher intelligibility scores than others.