A longitudinal study of very young children's vowel production.

PURPOSE: Ecologically realistic, spontaneous, adult-directed, longitudinal speech data of young children were described by acoustic analyses. METHOD: The first 2 formant frequencies of vowels produced by 6 children from different American English dialect regions were analyzed from ages 18 to 48 months. The vowels were from largely conversational contexts and were classified according to dictionary pronunciation. RESULTS: Within-subject formant frequency variability remained relatively constant for the span of ages studied. It was often difficult to detect overall decreases in the first 2 formant frequencies between ages 30 and 48 months. A study of the movement of the corner vowels with respect to the vowel centroid showed that the shape of the vowel space remained qualitatively constant from 30 through 48 months. CONCLUSIONS: The shape of the vowel space is established early in life. Some aspects of regional dialect were observed in some of the subjects at 42 months of age. The present study adds to the existing data on the development of vowel spaces by describing ecologically realistic speech.

Implications of peripheral muscular and anatomical development for the acquisition of lingual control for speech production: a review.

OBJECTIVES: Normally developing children learn to produce intelligible speech during rapid, non-uniform growth of their articulators and other vocal tract structures. The purpose of this review is to focus attention on the consequences of peripheral growth and development for the acquisition of lingual control for speech production. This paper (1) reviews physiological underpinnings of tongue shaping and movements that are likely to be changing in young children; (2) estimates, from previously published studies, the net consequences of growth of multiple vocal tract structures on lingual control; (3) integrates our findings with the example of [R] production, and (4) highlights areas where further investigations would be most helpful. PATIENTS AND METHODS: The authors searched the literature, including the PubMed database, for studies of the development of muscle proteins, muscle fibers, and motor units of the tongue, and of the growth of the tongue, jaw, adenoids, soft and hard palates, oral and pharyngeal cavities, and the vocal tract as a whole. CONCLUSIONS: Substantial anatomical and muscular data sets focused on children from 1-4 years of age, and rigorous definitions of the tongue boundaries are needed.

Source-tract interaction with prescribed vocal fold motion.

An equation describing the time-evolution of glottal volume velocity with specified vocal fold motion is derived when the sub- and supra-glottal vocal tracts are present. The derivation of this Fant equation employs a property explicated in Howe and McGowan [(2011) J. Fluid Mech. 672, 428-450] that the Fant equation is the adjoint to the equation characterizing the matching conditions of sub- and supra-glottal Green's functions segments with the glottal segment. The present aeroacoustic development shows that measurable quantities such as input impedances at the glottis, provide the coefficients for the Fant equation when source-tract interaction is included in the development. Explicit expressions for the Green's function are not required. With the poles and zeros of the input impedance functions specified, the Fant equation can be solved. After the general derivation of the Fant equation, the specific cases where plane wave acoustic propagation is described either by a Sturm-Liouville problem or concatenated cylindrical tubes is considered. Simulations show the expected skewing of the glottal volume velocity pulses depending on whether the fundamental frequency is below or above a sub- or supra-glottal formant. More complex glottal wave forms result when both the first supra-glottal fundamental frequencies are high and close to the first sub-glottal formant.

A study of high front vowels with articulatory data and acoustic simulations.

The purpose of this study is to test a methodology for describing the articulation of vowels. High front vowels are a test case because some theories suggest that high front vowels have little cross-linguistic variation. Acoustic studies appear to show counterexamples to these predictions, but purely acoustic studies are difficult to interpret because of the many-to-one relation between articulation and acoustics. In this study, vocal tract dimensions, including constriction degree and position, are measured from cinéradiographic and x-ray data on high front vowels from three different languages (North American English, French, and Mandarin Chinese). Statistical comparisons find several significant articulatory differences between North American English /i/ and Mandarin Chinese and French /i/. In particular, differences in constriction degree were found, but not constriction position. Articulatory synthesis is used to model the acoustic consequences of some of the significant articulatory differences, finding that the articulatory differences may have the acoustic consequences of making the latter languages' /i/ perceptually sharper by shifting the frequencies of F(2) and F(3) upwards. In addition, the vowel /y/ has specific articulations that differ from those for /i/, including a wider tongue constriction, and substantially different acoustic sensitivity functions for F(2) and F(3).

Analyses of vocal tract cross-distance to area mapping: an investigation of a set of vowel images.

Traditional models of mappings from midsagittal cross-distances to cross-sectional areas use only local cross-distance information. These are not the optimal models on which to base the construction of a mapping between the two domains. This can be understood because phonemic identity can affect the relation between local cross-distance and cross-sectional area. However, phonemic identity is not an appropriate independent variable for the control of an articulatory synthesizer. Two alternative approaches for constructing cross-distance to area mappings that can be used for articulatory synthesis are presented. One is a vowel height-sensitive model and the other is a non-parametric model called loess. These depend on global cross-distance information and generally perform better than the traditional models.

Sagittal area of the vocal tract in young female children.

OBJECTIVE: To measure the sagittal areas of the front and back cavities of the vocal tract in children acquiring speech. PATIENTS AND METHODS: Ten female children were selected from the Serial Experimental collection of the Burlington Growth Centre in Toronto, Canada. Each of the 10 children was seen annually from ages 3 through 8. Data collections included lateral cephalograms in occlusion. We traced those cephalograms and identified landmarks to delineate the front and back cavities. The sagittal areas of the front and back cavities were calculated. A measure of the angle of the head to the cervical vertebrae was made. RESULTS: Front cavities were larger and grew faster. For both front and back cavities, age, angle measure, and the interaction of age and angle measure were significant. CONCLUSION: Space available for the tongue to maneuver is greater anteriorly than posteriorly even when the jaw is maximally elevated.

Comments on single-mass models of vocal fold vibration.

Proposed mechanisms for single-mass oscillation in the vocal tract are examined critically. There are two areas that distinguish single-mass models: in the sophistication of the air flow modeling near the oscillator and whether or not oscillation depends on acoustic feedback. Two recent models that do not depend on acoustic feedback are examined in detail. One model that depends on changing flow separation points is extended with approximate calculations.

Influence of the ventricular folds on a voice source with specified vocal fold motion.

The unsteady drag on the vocal folds is the major source of sound during voiced speech. The drag force is caused by vortex shedding from the vocal folds. The influence of the ventricular folds (i.e., the "false" vocal folds that protrude into the vocal tract a short distance downstream of the glottis) on the drag and the voice source are examined in this paper by means of a theoretical model involving vortex sheets in a two-dimensional geometry. The effect of the ventricular folds on the output acoustic pressure is found to be small when the movement of the vocal folds is prescribed. It is argued that the effect remains small when fluid-structure interactions account for vocal fold movement. These conclusions can be justified mathematically when the characteristic time scale for change in the velocity of the glottal jet is large compared to the time it takes for a vortex disturbance to be convicted through the vocal fold and ventricular fold region.

A method for finding constrictions in high front vowels.

The purpose of this study was to devise a consistent and robust method for defining vocal tract constrictions in high front vowels. A procedure was devised to find the length and position of the articulatory constriction in high front vowels that is not sensitive to local fluctuations in vocal tract shape and to the constriction-defining parameters. A method based on a visual examination of plots for constriction length and position as functions of the constriction-defining parameters was found to provide stable constriction definitions.

Acoustic-articulatory mapping in vowels by locally weighted regression.

A method for mapping between simultaneously measured articulatory and acoustic data is proposed. The method uses principal components analysis on the articulatory and acoustic variables, and mapping between the domains by locally weighted linear regression, or loess [Cleveland, W. S. (1979). J. Am. Stat. Assoc. 74, 829-836]. The latter method permits local variation in the slopes of the linear regression, assuming that the function being approximated is smooth. The methodology is applied to vowels of four speakers in the Wisconsin X-ray Microbeam Speech Production Database, with formant analysis. Results are examined in terms of (1) examples of forward (articulation-to-acoustics) mappings and inverse mappings, (2) distributions of local slopes and constants, (3) examples of correlations among slopes and constants, (4) root-mean-square error, and (5) sensitivity of formant frequencies to articulatory change. It is shown that the results are qualitatively correct and that loess performs better than global regression. The forward mappings show different root-mean-square error properties than the inverse mappings indicating that this method is better suited for the forward mappings than the inverse mappings, at least for the data chosen for the current study. Some preliminary results on sensitivity of the first two formant frequencies to the two most important articulatory principal components are presented.

Speech production in 12-month-old children with and without hearing loss.

PURPOSE: The purpose of this study was to compare speech production at 12 months of age for children with hearing loss (HL) who were identified and received intervention before 6 months of age with those of children with normal hearing (NH). METHOD: The speech production of 10 children with NH was compared with that of 10 children with HL whose losses were identified (better ear pure-tone average at 0.5, 1, and 2 kHz poorer than 50 dB HL) and whose intervention started before 6 months of age. These children were recorded at 12 months of age interacting with a parent. Three properties of speech production were analyzed: (a) syllable shape, (b) consonant type, and (c) vowel formant frequencies. RESULTS: Children with HL had (a) fewer multisyllable utterances with consonants, (b) fewer fricatives and fewer stops with alveolar-velar stop place, and (c) more restricted front-back tongue positions for vowels than did the children with NH. CONCLUSION: Even when hearing loss is identified shortly after birth, children with HL do not develop speech production skills as their peers with NH do at 12 months of age. This suggests that researchers need to consider their approaches to early intervention carefully.

Predicting midsagittal pharyngeal dimensions from measures of anterior tongue position in Swedish vowels: statistical considerations.

In a re-analysis of x rays of speakers producing Swedish vowels, midsagittal pharyngeal dimensions were predicted from anterior tongue positions using procedures based on estimated tongue pellet positions. Principal component analysis was used to reduce the number of pellet degrees of freedom from eight to three prior to applying linear regression from these three independent variables to dependent vocal tract midsagittal cross distances. Except for the regions around the laryngopharynx and uvula, the pharynx dimensions were predictable from linear regressions and were significant at the p<0.05 level. Numerical experiments show that it is crucial to reduce the number of independent variables in tests of statistical significance.

Perception of synthetic vowel exemplars of 4-year-old children and estimation of their corresponding vocal tract shapes.

Formant scalings for vowel exemplars of American 4 year olds who were imitating adult production were used along with published data of American adult male vowel production to synthesize /a, ae, u, i/. Other vowel exemplars were also synthesized. Adult listeners were asked to categorize these synthetic vowels in a forced choice task. With some exceptions, the formant frequencies preferred for the vowels /a, ae, u, i/ were close to the published data. In order to gain insight on children's articulation during imitation of vowels /a, ae, u, i/, a five-tube model was used in an algorithm to infer vocal tract shape from the first three formant frequencies of the adult productions, the formant frequencies derived for 4 year olds by scaling, and formant frequencies for 4 year olds derived based on the listening experiments. It was found that the rear tube length for the children, in proportionate terms, was nearly always greater than that of the adult. The rear tube length was proportionately twice as long in children compared to adults for the vowel /u/. Tongue root flexibility and the oblique angle between the pharynx and mouth may be more important than pharynx length in determining formant scalings for 4 year old children.

Secant lines in a preliminary study of phonetic reduction of /I/ and /epsilon/.

Two subjects from the X-Ray Microbeam Speech Production Database were examined in their production of the vowels /I/ and /epsilon/ in alveolar and dental consonant contexts. Secant lines, or first-order splines, between the three most anterior pellets were examined at vowel critical times. These critical times were zero crossings in the tangential acceleration of the midpoints of the secant lines. We expected and found, in general, that vowel reduction occurred as a function of vowel duration in measures of the secant line midpoint-to-palate distance and secant line orientation at vowel critical times. The shorter the vowel, the smaller the distance of the secant line midpoints to the palate and the less downward the orientation of the secant lines at the vowel critical times. Phonetic reduction was also apparent in the formant frequencies. There were differences between the speakers in terms of the range of vowel duration and degree of reduction. The subjects differed in the functional parts of the tongue spanned by the secant lines and the shape of their palates. These differences were factors in the observed relations between formant frequencies and the articulatory, secant line measures for each subject.

Rational approximations of viscous losses in vocal tract acoustic modeling.

The modeling of viscous losses in acoustic wave transmission through tubes by a boundary layer approximation is valid if the thickness of the boundary layer is small compared to the hydraulic radius. A method was found to describe the viscous losses that extends the frequency range of the model to very low frequencies and very thin tubes. For higher frequencies, this method includes asymptotically the spectral effects of the boundary layer approximation. The method provides a simplification for the rational approximation of the spectral effects of viscous losses.

Development of [j] in young, midwestern, American children.

Beginning at the age of about 14 months, eight children who lived in a rhotic dialect region of the United States were recorded approximately every 2 months interacting with their parents. All were recorded until at least the age of 26 months, and some until the age of 31 months. Acoustic analyses of speech samples indicated that these young children acquired [inverted r] production ability at different ages for [inverted r]'s in different syllable positions. The children, as a group, had started to produce postvocalic and syllabic [inverted r] in an adult-like manner by the end of the recording sessions, but were not yet showing evidence of having acquired prevocalic [inverted r]. Articulatory limitations of young children are posited as a cause for the difference in development of [inverted r] according to syllable position. Specifically, it is speculated that adult-like prevocalic [inverted r] production requires two lingual constrictions: one in the mouth, and the other in the pharynx, while postvocalic and syllabic [inverted r] requires only one oral constriction. Two lingual constrictions could be difficult for young children to produce.