How Do Children Organize Their Speech in the First Years of Life? Insight From Ultrasound Imaging

Article (PDF Available)inJournal of Speech Language and Hearing Research 61(6):1355-1368 · June 2018with 384 Reads
DOI: 10.1044/2018_JSLHR-S-17-0148
Cite this publication
Abstract
(in press; excerpt from a non final abstract due to copyright restriction) This study reports on a cross-sectional investigation of lingual coarticulation in 57 typically developing German children (four cohorts from 3.5 to 7 years of age) as compared with 12 adults. It examines whether the organization of lingual gestures for intrasyllabic coarticulation differs as a function of age and consonantal context. Using the technique of ultrasound imaging, we recorded movement of the tongue articulator during the production of pseudo words including various vocalic and consonantal contexts. Results from linear mixed effects models show greater lingual coarticulation in all groups of children as compared to adults with a significant decrease from the kindergarten years (at 3; 4; 5) to the end of the first year into primary school (at 7). Additional differences in coarticulation degree were found across and within age groups as a function of the onset consonant identity (/b/, /d/ and /g/). Results support the view that although coarticulation degree decreases with age, children do not organize consecutive articulatory gestures with a uniform organizational scheme (e.g., segmental or syllabic). Instead results suggest coarticulatory organization is sensitive to the underlying articulatory properties of the segments combined.
Figures - uploaded by Aude Noiray
Author content
All content in this area was uploaded by Aude Noiray
Content may be subject to copyright.
No caption available
… 
This manuscript is not for public dissemination. It is published in its final form in Journal of Speech, Language and
Hearing Research, June, 2018. https://jslhr.pubs.asha.org/article.aspx?articleid=2681858
!
!
1!
1
!
2
!
3
!
4
!
How do children organize their speech in the first years of life?
5
!
Insight from ultrasound imaging
6
!
7
!
Aude Noiray12, Dzhuma Abakarova1, Elina Rubertus1, Stella Krüger1, Mark Tiede 2
8
!
9
!
1 Laboratory for Oral Language Acquisition, University of Potsdam, Germany
10
!
2 Haskins Laboratories, New Haven, USA
11
!
12
!
This research was supported by a grant from the Deutsche Forschungsgemeinschaft
13
!
14
!
15
!
16
!
17
!
18
!
19
!
20
!
21
!
22
!
Correspondence concerning this article should be addressed to Aude Noiray, University of
23
!
Potsdam (Germany).
24
!
Contact: anoiray@uni-potsdam.de
25
!
26
!
27
!
28
!
29
!
30
!
31
!
32
!
33
!
34
!
This manuscript is not for public dissemination. It is published in its final form in Journal of Speech, Language and
Hearing Research, June, 2018. https://jslhr.pubs.asha.org/article.aspx?articleid=2681858
!
!
2!
1
!
ABSTRACT
2
!
Purpose: This study reports on a cross-sectional investigation of lingual coarticulation in 57
3
!
typically developing German children (four cohorts from 3.5 to 7 years of age) as compared with
4
!
12 adults. It examines whether the organization of lingual gestures for intrasyllabic coarticulation
5
!
differs as a function of age and consonantal context.
6
!
Method: Using the technique of ultrasound imaging, we recorded movement of the tongue
7
!
articulator during the production of pseudo words including various vocalic and consonantal
8
!
contexts.
9
!
Results: Results from linear mixed effects models show greater lingual coarticulation in all groups
10
!
of children as compared to adults with a significant decrease from the kindergarten years (at 3; 4;
11
!
5) to the end of the first year into primary school (at 7). Additional differences in coarticulation
12
!
degree were found across and within age groups as a function of the onset consonant identity (/b/,
13
!
/d/ and /g/).
14
!
Conclusions: Results support the view that although coarticulation degree decreases with age,
15
!
children do not organize consecutive articulatory gestures with a uniform organizational scheme
16
!
(e.g., segmental or syllabic). Instead results suggest coarticulatory organization is sensitive to the
17
!
underlying articulatory properties of the segments combined.
18
!
19
!
Keywords: Children, Development, Language, Speech motor control, Speech production
20
!
21
!
22
!
23
!
24
!
25
!
26
!
27
!
28
!
29
!
30
!
31
!
32
!
33
!
This manuscript is not for public dissemination. It is published in its final form in Journal of Speech, Language and
Hearing Research, June, 2018. https://jslhr.pubs.asha.org/article.aspx?articleid=2681858
!
!
3!
How do children organize their speech in the first years of life?
1
!
Insight from ultrasound imaging
2
!
3
!
INTRODUCTION
4
!
5
!
In the domain of spoken language acquisition, great attention has been focused on
6
!
coarticulation, which concerns the overlapping of articulatory gestures for neighboring segments
7
!
(for a review see Hardcastle & Hewlett, 2006). Coarticulation is a fundamental characteristic of
8
!
fluent speech. It is an important mechanism to investigate as it taps into the phonetic instantiations
9
!
of phonological units from various sizes such as phonemes or syllables and therefore offers a
10
!
chance to reveal how unit organization matures over time as children learn to speak their native
11
!
language. In addition, coarticulation engages multiple speech articulators (e.g., the lips, the
12
!
tongue) whose actions must be coordinated in time and in the space of the vocal tract to produce
13
!
intelligible phonetic outputs in the native language. Investigating the development of
14
!
coarticulatory patterns therefore provides a unique opportunity to address both the maturation of
15
!
the speech motor system and its attunement to the phonetic regularities of the language spoken.
16
!
In this study, we were specifically interested in examining how differences in lingual vowel-
17
!
to-consonant coarticulation can shed light on the phonetic organization of speech in young
18
!
German children, from 3 years of age (when they are in kindergarten) to 7 years of age when
19
!
children are in primary school. In addition, we aimed to provide a first quantitative survey of
20
!
anticipatory coarticulation in German learners. Much developmental work on lingual
21
!
coarticulation has focused on English variants. However, studies in languages other than English
22
!
are needed to possibly disentangle universal versus language specific patterns of coarticulation. In
23
!
German at least, most assessments of consonant acquisition have used measures of individual
24
!
production accuracy (e.g., Fox-Boyer, 2006). Using the technique of ultrasound imaging, we
25
!
examined the organization of gestures of the tongue, an organ whose control is essential to
26
!
vowels’ and consonants’ acquisition (e.g., Barbier, Perrier, Ménard, & Payan, 2015; Klein et al.
27
!
2013; Ménard & Noiray, 2011; Noiray, Ménard, & Iskarous, 2013; Rubertus, Abakarova, Ries, &
28
!
Noiray, 2016; Song, Demuth, Shattuck-Hufnagel, & Ménard, 2013; Zharkova, Hewlett,
29
!
Hardcastle, & Lickley, 2014). In the past, articulatory tracking methods such as electromagnetic
30
!
articulography (EMA) and electropalatography (EPG) have been employed in school-aged
31
!
children and adolescents (e.g., EPG: Cheng, Murdoch, Goozée, & Scott, 2007; Gibbon, & Wood,
32
!
2010; EMA: Katz & Bharadwaj 2001; Terband, Maassen, Lieshout, & Nijland, 2011). More
33
!
This manuscript is not for public dissemination. It is published in its final form in Journal of Speech, Language and
Hearing Research, June, 2018. https://jslhr.pubs.asha.org/article.aspx?articleid=2681858
!
!
4!
recently, the technique of ultrasound imaging has been adapted to the developmental field to make
1
!
articulatory recordings of the tongue possible in young populations (e.g., Barbier, Perrier, Ménard,
2
!
& Payan, 2015; Ménard & Noiray, 2011; Noiray, Ménard, & Iskarous, 2013; Song, et al., 2013;
3
!
Zharkova, 2017). Compared to EMA and EPG, ultrasound is a more suitable technique to use with
4
!
young children because it does neither require long preparation time prior to testing nor invasive
5
!
procedures to track tongue movement during speech (e.g., gluing EMA pellets on young
6
!
children´s tongue or placing an artificial palate). Hence, with the technique of ultrasound imaging,
7
!
it is now possible to revisit questions related to speech organization in young children while
8
!
directly examining the articulatory mechanisms underlying speech production rather than inferring
9
!
those mechanisms from the acoustic outputs.
10
!
11
!
What are the units of speech production in children?
12
!
Finding the units of speech organization in the first years of life has been one of the most
13
!
challenging endeavors for developmental psycholinguists, but the quest is important for advancing
14
!
both theories of language acquisition and clinical assessment of disordered speech.
15
!
Over the past two decades, research examining intra-syllabic coarticulatory patterning in typically
16
!
developing children has provided conflicting results and hypotheses regarding the nature of these
17
!
units. A number of studies have reported less coarticulation in children compared to adults with
18
!
limited influence of the vowel on the preceding consonant. Such findings lead to the hypothesis
19
!
that spoken language organization is initially segmentally driven (e.g., Gibson & Ohde, 2007;
20
!
Green, Moore, & Reilly, 2002; Katz, Kripke, & Tallal, 1991; Kent, 1983). In this view, children
21
!
are supposed to proceed through a sequential maturation process by which articulatory controls for
22
!
individual segments progressively develop into more complex inter-articulator organizations for
23
!
larger units, with increasing intra-syllabic coarticulation as a result. An opposite view holds that
24
!
children initially display greater consonant-vowel (CV) coarticulation than adults, suggesting a
25
!
broader planning unit of their speech than the segmental unit (e.g., Goodell, & Studdert-Kennedy,
26
!
1993; Nijland et al. 2002; Nittrouer & Whalen, 1989; Nittrouer, Studdert-Kennedy, & Neely,
27
!
1996; Rubertus et al., 2015). In this more holistic perspective, maturation of coarticulatory
28
!
patterns would consist in decreasing encroachment between consonantal and vocalic components
29
!
and development of increasingly differentiated controls over individual articulators for a more
30
!
segmental organization of articulatory movements. Other studies have found equivalent patterns
31
!
of coarticulatory degree in children and adults but reported greater variability in children’s
32
!
patterns (e.g., Katz, Kripke, & Tallal, 1991; Munson, 2004; Repp, 1986; Sereno, Baum, Marean,
33
!
This manuscript is not for public dissemination. It is published in its final form in Journal of Speech, Language and
Hearing Research, June, 2018. https://jslhr.pubs.asha.org/article.aspx?articleid=2681858
!
!
5!
& Lieberman, 1987). To date, organizational units of speech production are still discussed (see
1
!
satellite workshop in Laboratory Phonology 15, July 2016 dedicated to this topic).
2
!
Additional research is evidently needed not only to disentangle the origin(s) of current
3
!
theoretical discrepancies; but also because a detailed understanding of lingual coarticulatory
4
!
development over age in typically developing children (TDs) would provide useful information
5
!
for advancing detection of atypical trajectories (Maas & Mailend, 2017). Indeed, a series of
6
!
experimental studies conducted by Nijland and colleagues revealed inconsistent coarticulatory
7
!
organization in children with developmental apraxia of speech (DAS) compared to TD controls
8
!
(e.g., Nijland, Maassen, & van der Meulen, 2003). While some children with DAS seem to exhibit
9
!
greater coarticulation than TDs, others show the opposite patterns. More recently, Terband (2017)
10
!
examined coarticulatory patterns from 16 children with DAS aged between 5.5 and 7.5 years
11
!
producing /bi, di, bu, du/ and reported greater coarticulation than the 8 age-matched TDs tested for
12
!
comparison. However, deviant coarticulatory patterns in children with DAS were only observed in
13
!
some phonetic contexts but not all. This suggests that the deficit in anticipatory coarticulation
14
!
observed in children with DAS is not uniform as assumed in the ASHA descriptions (ASHA,
15
!
2007) but specific to certain phoneme combinations that may involve more complex articulatory
16
!
coordinations compared to others. Difficulty in coarticulatory organization has also been noticed
17
!
in stuttering children (e.g., Soo-Eun, Ohde, & Conture, 2002) who seem to exhibit smaller
18
!
coarticulatory differences across consonantal contexts than typically developing age-matched
19
!
children. Taken together, these results have important implications as to the links between the
20
!
articulatory properties of the speech material investigated, speech motor control and the breadth of
21
!
coarticulatory organization in atypical development. To provide reference data in German, this
22
!
study focuses on lingual vowel-to-consonant coarticulation in typically developing children.
23
!
!
24
!
Differences in lingual coarticulation degree and resistance across consonantal contexts
25
!
An important variable to consider when investigating variance in lingual coarticulatory
26
!
patterns within a sample of participants or across populations regards the articulatory properties of
27
!
the sequences produced. In adults, differences in intra-syllabic coarticulation degree within
28
!
individuals reflect differences in consonants’ place of articulation with labial-V syllables showing
29
!
a high coarticulation degree (CD) contrary to alveolar or alveopalatal stop-V syllables, which
30
!
show a lower degree of coarticulation between consonantal and vocalic lingual gestures.
31
!
Interestingly, in adults these patterns have been consistently reported across various languages
32
!
(e.g., in American English: Fowler, 1994; Fowler & Brancazio, 2001; Iskarous et al., 2011;
33
!

Supplementary resources

  • Article
    Full-text available
    This study examines the temporal organization of vocalic anticipation in German children from 3 to 7 years of age and adults. The main objective was to test for non-linear processes in vocalic anticipation, which may result from the interaction between lingual gestural goals for individual vowels, and those for their neighbors over time. The technique of ultrasound imaging was employed to record tongue movement at five time points throughout short utterances of the form V1#CV2. Vocalic anticipation was examined with Generalized Additive Modeling, an analytical approach allowing for the estimation of both linear and non-linear influences on anticipatory processes. Both adults and children exhibit non-linear patterns of vocalic anticipation over time with the degree and extent of vocalic anticipation varying as a function of the individual consonants and vowels assembled. However, noticeable developmental discrepancies were found with vocalic anticipation being present earlier in children´s utterances at 3-4-5 years of age in comparison to adults and to some extent 7-year-old children. A narrowing of speech production organization from large chunks in kindergarten to more segmentally differentiated and contextually-specified organizations from kindergarten to primary school to adulthood, although variation in the temporal overlap of lingual gestures for consecutive segments is already present in the youngest cohorts. In adults, non-linear anticipatory patterns over time suggest a strong differentiation between the gestural goals for consecutive segments. In children, this differentiation is not yet mature: vowels show greater prominence over time and seem activated more in-phase with those of previous segments relative to adults.
  • Article
    Full-text available
    This study aims at evaluating speech motor skills in the fluent speech of a cohort of stuttering Italian children. Stuttering is a neurodevelopmental disorder that may arise from an innate limitation of the speech motor control system, which fails to prepare and organize the movements required for fluent speech (Van Lieshout, Hulstijn, & Peters, 2004). Anticipatory coarticulation in CV sequences and stability of speech movements have been used as measures of the maturity of articulatory processes for fluent speech production. This study aims to assess if direct measures of speech dynamics can identify impaired mechanisms in stuttering speech during a phrase-repetition task. The Ultrasound Tongue Imaging data of eight school-aged children, half of whom stutter while the other half don’t, show different articulatory patterns between the two groups, for both motor aspects under investigation. Articulatory data show that the stuttering group presents a higher degree of intra-syllabic coarticulation compared to the control group and decreased stability (i.e. more variability) through multiple repetitions of the same alveolar and velar item. Outcomes of this study suggest that the speech motor control system of children who stutter is less mature in preparing and executing the speech gestures required for fluent speech. This study contributes to shedding light on the impaired articulatory patterns involved in stuttering speech and to identifying the diagnostic markers of the disorder by evaluating the speech of children close to the onset of stuttering.
  • Article
    Full-text available
    In the first years of life, children differ greatly from adults in the temporal organization of their speech gestures in fluent language production. However, dissent remains as to the maturational direction of such organization. The present study sheds new light on this process by tracking the development of anticipatory vowel-to-vowel coarticulation in a cross-sectional investigation of 62 German children (from 3.5 to 7 years of age) and 13 adults. It focuses on gestures of the tongue, a complex organ whose spatiotemporal control is indispensable for speech production. The goal of the study was threefold: 1) investigate whether children as well as adults initiate the articulation for a target vowel in advance of its acoustic onset, 2) test if the identity of the intervocalic consonant matters and finally, 3) describe age-related developments of these lingual coarticulatory patterns.
  • Preprint
    Full-text available
    A conceptual framework for modeling articulatory control is presented in a tutorial fashion. The framework incorporates three mechanisms: selection, coordination, and intention. Selection is mechanism for governing the choice and ordering of articulatory movements, and operates through an activation code. Coordination is a mechanism for governing the control of movement timing in precise manner, and operates through a phase code regulated by coupled oscillators. Intention is a mechanism for determining the target state of the vocal tract, and operates through a spatial code derived from integrating over parameter fields. All three mechanisms are inherently dynamic, and their interactions provide a basis for understanding a wide variety of phonetic and phonological patterns in development and across languages. An integrated model of the mechanisms is described and applied to a range of empirical phenomena.
  • Article
    Full-text available
    (In press. Pre-final version, due to copyright) In previous research, Mutual Information (MI) was employed to quantify the physical information shared between consecutive phonological segments, based on electromagnetic articulography data (EMA). In this study, MI is extended to quantifying coarticulatory resistance (CR) versus overlap in German adult speakers using ultrasound imaging. Two measurements are tested as input to MI: 1) the highest point on the tongue body and, 2) the first coefficient of the discrete Fourier transform (DFT) of the whole tongue contour.
  • Book
    This monograph arose from a conference on the Production of Speech held at the University of Texas at Austin on April 28-30, 1981. It was sponsored by the Center for Cognitive Science, the College of Liberal Arts, and the Linguistics and Psychology Departments. The conference was the second in a series of conferences on human experimental psychology: the first, held to commemorate the 50th anniversary of the founding of the Psychology Department, resulted in publication of the monograph Neural Mechanisms in Behavior, D. McFadden (Ed.), Springer-Verlag, 1980. The choice of the particular topic of the second conference was motivated by the belief that the state of knowledge of speech production had recently reached a critical mass, and that a good deal was to be gained from bringing together the foremost researchers in this field. The benefits were the opportunity for the participants to compare notes on their common problems, the publication of a monograph giving a comprehensive state-of-the-art picture of this research area, and the provision of enormous intellectual stimulus for local students of this topic.
  • Article
    Visual biofeedback is the process of gaining awareness of physiological functions through the display of visual information. As speech is concerned, visual biofeedback usually consists in showing a speaker his/her own articulatory movements, which has proven useful in applications such as speech therapy or second language learning. This article presents a novel method for automatically animating an articulatory tongue model from ultrasound images. Integrating this model into a virtual talking head enables to overcome the limitations of displaying raw ultrasound images, and provides a more complete and user-friendly feedback by showing not only the tongue, but also the palate, teeth, pharynx, etc. Altogether, these cues are expected to lead to an easier understanding of the tongue movements. Our approach is based on a probabilistic model which converts raw ultrasound images of the vocal tract into control parameters of the articulatory tongue model. We investigated several mapping techniques such as the Gaussian Mixture Regression (GMR), and in particular the Cascaded Gaussian Mixture Regression (C-GMR) techniques, recently proposed in the context of acoustic-articulatory inversion. Both techniques are evaluated on a multispeaker database. The C-GMR consists in the adaptation of a GMR reference model, trained with a large dataset of multimodal articulatory data from a reference speaker, to a new source speaker using a small set of adaptation data recorded during a preliminary enrollment session (system calibration). By using prior information from the reference model, the C-GMR approach is able (i) to maintain good mapping performance while minimizing the amount of adaptation data (and thus limiting the duration of the enrollment session), and (ii) to generalize to articulatory configurations not seen during enrollment better than the GMR approach. As a result, the C-GMR appears to be a good mapping technique for a practical system of visual biofeedback.
  • Article
    Purpose: The purpose of this study was, first, to expand our understanding of typical speech development regarding segmental contrast and anticipatory coarticulation, and second, to explore the potential diagnostic utility of acoustic measures of fricative contrast and anticipatory coarticulation in children with speech sound disorders (SSD). Method: In a cross-sectional design, 10 adults, 17 typically developing children, and 11 children with SSD repeated carrier phrases with novel words with fricatives (/s/, /ʃ/). Dependent measures were 2 ratios derived from spectral mean, obtained from perceptually accurate tokens. Group analyses compared adults and typically developing children; individual children with SSD were compared to their respective typically developing peers. Results: Typically developing children demonstrated smaller fricative acoustic contrast than adults but similar coarticulatory patterns. Three children with SSD showed smaller fricative acoustic contrast than their typically developing peers, and 2 children showed abnormal coarticulation. The 2 children with abnormal coarticulation both had a clinical diagnosis of childhood apraxia of speech; no clear pattern was evident regarding SSD subtype for smaller fricative contrast. Conclusions: Children have not reached adult-like speech motor control for fricative production by age 10 even when fricatives are perceptually accurate. Present findings also suggest that abnormal coarticulation but not reduced fricative contrast is SSD-subtype-specific. Supplemental materials: S1: https://doi.org/10.23641/asha.5103070. S2 and S3: https://doi.org/10.23641/asha.5106508.
  • Conference Paper
    Full-text available
    A successful characterization of vocal tract control during speech needs to account for regular variability in the degree of coarticulatory overlap allowed by different speech segments. While some segments allow for large degree of articulatory overlap, others show high coarticulation resistance (CR) i.g. ability to resist influence from neighbors and retain control over articulators across contexts. Despite the importance of the CR phenomenon for theories and models of speech production, a unified quantitative measure of coarticulation resistance has not been established yet. The most prominent description of CR, DAC scale (Recasens & Espinosa, 2009), has some limitations: first, it subjectively divides segments into several categories based exclusively on the degree of lingual coarticulation they exhibit across contexts. Second, this categorization is based on the measures of coarticulation that quantify only linear dependencies. Generally, a wide variety of experimental techniques and corresponding quantification methods make it difficult to directly compare CR estimates across studies. Recently, Iskarous et al. (2013) suggested measuring CR with Mutual Information (MI), or the amount of information shared by a given segment with other segments across contexts. The MI is non-parametric method that does not make assumptions about distribution but rather estimates it from data. The MI values have been shown by Iskarous et al (2013) to capture the CR effects of place and manner, as well as time differences in CR. In our study we extend the application of MI to quantifying coarticulation from ultrasound images. We investigate the effect of consonantal context on vowel-to-consonant coarticulation in German adults by looking at the position of the highest point of the tongue. Our results show that CR of German consonants exhibits the pattern /z > d > g > b/ in horizontal dimension and /z > g > d > b/ in the vertical dimension. These finding corroborates those made by Iskarous et al (2013) for German voiceless consonants for tongue body articulator using EMA data. This suggests that MI measure allows for cross-methodological comparisons and generalizations of quantitative findings. Temporal aspects of coarticulation resistance for different segments are currently being investigated by comparing MI values at different time points during consonant production. We are also applying MI to quantify whole tongue contours to compare aspects of coarticulation captured by different methods of tongue shape quantification.
  • Article
    Full-text available
    In this study, vowel-on-consonant lingual coarticulation at [t] closure offset was compared in 5-year-old children and 13-year-old adolescents. The study aimed to establish whether, by the end of the closure, children from the younger age group adjust the tongue shape to the following vowels to the same extent as adolescents. Ten 5-year-olds and ten 13-year-olds, all speakers of Scottish Standard English, produced [t]-vowel syllables with the vowels [i] and [a], in a carrier phrase. Measures of tongue shape based on midsagittal ultrasound imaging data were used to compare anticipatory coarticulation and within-speaker variability across groups. Both age groups changed the extent of tongue dorsum bunching in order to coarticulate the consonant with the following vowels. The 5-year-old children, unlike the adolescents, did not consistently modify the bunching location within the tongue curve to accommodate the tongue shape to that of the upcoming vowel. Token-to-token variability was significantly greater in the younger age group. The results suggest that vowel-on-[t] coarticulatory patterns produced by typically developing children are affected by the development of motor control, with articulatory constraints on the tongue limiting the extent of lingual coarticulation in 5-year-old children. The findings on typical coarticulation development are relevant for clinical practice, and they highlight the need for more detailed descriptions of how phonetic characteristics of speech sounds affect coarticulation throughout childhood.
  • Article
    A new method for quantifying contextual variability at different regions of the tongue using ultrasound spline data reveals that tongue body coarticulatory resistance for Catalan consonants and vowels in VCV sequences decreases in the progression [ʎ, ɲ, ʃ] > [s, r] > [l, ɾ, t, n] > [ð] and [i, e] > [a] > [o] > [u]. These consonant and vowel hierarchies support the degree of articulatory constraint model of coarticulation according to which coarticulatory resistance depends on whether a given lingual region is involved in the formation of a closure or constriction and on the severity of the manner of articulation requirements. Data show that this coarticulatory scenario holds not only at the palatal zone, as revealed by previous coarticulation studies, but at the velar and pharyngeal zones as well. Partial exceptions are [s] and [i], which may allow for some more contextual variability than expected at the back of the vocal tract. Another major finding is that tongue body coarticulatory resistance and aggressiveness are highly positively correlated. The implications of these experimental results for speech production organization and sound change are discussed.
  • Conference Paper
    Full-text available
    We present here a customized method developed jointly by scientists at LOLA (Potsdam University) and Haskins Laboratories (New Haven) for the recording of both tongue and lip motion during speech tasks in young children. The method is currently being used to investigate the development of 1) coarticulation (resistance and anticipatory coarticulation, cf. two other abstracts submitted); and 2) articulatory coordination in preschoolers compared with adults who have mature control of their speech production system. Children are recorded with a portable ultrasound system (Sonosite Edge, 48Hz) with a small probe fixed on a custom-made probe holder and ultrasound stand. The probe holder was specifically designed to allow for natural vertical motion of the jaw but prevent motion in the lateral and horizontal translations. The set up is integrated into a child-friendly booth that facilitates integrating the production tasks into games. Ultrasound video data are collected concurrently with synchronized audio recorded via a microphone (Shure, 48kHz,), pre-amplified before being recorded onto a desktop computer. In addition to tongue motion, a frontal video recording of the face is obtained with a camcorder (Sony HDR-CX740VE, fps: 50Hz). This video is used to track lip motion for subsequent labial measurements, and to track head and probe motion for transforming contours extracted from the ultrasound images to a head-based coordinate system. The speech signal is also recorded via the built-in camcorder microphone, and synchronization of both video signals (from the ultrasound and the camcorder) is performed through audio cross-correlation in post-processing. Lip motion is characterized with a video shape tracking system (Lallouache 1991) previously used for examining anticipatory coarticulation in adults (Noiray et al., 2011) and children (Noiray et al., 2004; 2008). During production tasks, the lips of our young participants are painted in blue as this color maximized contrast with the skin. In post-processing these blue shapes are then tracked for calculation of lip aperture, interolabial area and upper lip protrusion. Tongue contours derived from ultrasound are relative to the orientation of the probe with respect to the tongue surface. To correct for jaw displacement and (pitch) rotation of the head we compute two correction signals similar to the HOCUS method described in Whalen et al. (2005), but in this case derived from tracking the positions of blue reference dots in the video signal using custom Matlab procedures. The displacement of the probe relative to the centroid of dots placed on each speaker's forehead provides a vertical correction signal. The orientation of dots placed on the cheek observed within the video image through a mirror oriented at 45° giving a profile view provides a pitch rotation correction signal around the lateral axis. Application of these two signals to the extracted contours allows for their consistent comparison in a head-centric coordinate system.