Infants’ vocalizations analyzed with an articulatory model: A preliminary report (original) (raw)
Related papers
2005
This article describes the development of an auditory coding system and an associated web-based XML database to document and analyze speech development in infants from Canada, Morocco, France, and China. The XML database will allow researchers on this international research project, based at the Department of Linguistics, University of Victoria, to collaborate in data collection and analysis, facilitating a better understanding of phonetic development in infants from diverse language backgrounds. We hypothesize that infants from all language backgrounds employ laryngeal constriction as the primary means of phonetic development in the first months of life, using this experience as a base to learn the sounds of their native language in the second half of the first year. Preliminary results from English-speaking infants support this hypothesis.
Contingencies governing the production of fricatives, affricates, and liquids in babbling
Applied Psycholinguistics, 2000
Studies of early-developing consonants (stops, nasals, and glides) in babbling have shown that most of the variance in consonants and their associated vowels, both within and between syllables, is due to a “frame” produced by mandibular oscillation, with very little active contribution from intrasyllabic or intersyllabic tongue movements. In a study of four babbling infants, the prediction that this apparently basic “frame dominance” would also apply to late-developing consonants (fricatives, affricates, and liquids) was tested. With minor exceptions, confirming evidence for both the predicted intrasyllabic and intersyllabic patterns was obtained. Results provide further evidence for the frame dominance conception, but suggest that the early rarity of late-developing consonants may be primarily a result of intrasegmental production difficulty.
Acoustic evidence for the development of speech
Research Laboratory of Electronics, Massachusetts Institute of Technology eBooks, 1989
This thesis develops models of relationships between the physical attributes of young children and the acoustic characteristics of their vocalizations. Two questions are addressed. What are the acoustic characteristics of children's speech? How can acoustic data be used to describe the development of speech? Children's vocalizations exhibit a range of characteristics, becoming more adult-like as a child grows and develops abilities in motor control and cognitive functioning. Developmental changes which take place during childhood result from changes in the child's anatomy and in motor-control and cognitive abilities. The physical attributes of a child's vocal system and the constraints on motor control must be known in order to construct models of the characteristics of children's speech. These attributes include the dimensions of the lungs, trachea, glottis, vocal tract and head and the tissue properties of the glottal structure and vocal-tract walls. Data on physical attributes of the speech production systems of young children are summarized in the form of model parameters in this thesis. A new model of the fundamental frequency of vocal-fold vibratic: is developed; this model is based on the theory of bending beams. An important parameter of the model is the ratio of vocal-fold thickness to length. This ratio affects the fundamental frequencies which are calculated from the model. Predicted durations rely on assumptions of the volume of air used in speech and the airflow through the glottis and vocal tract. The predictions imply that children's voiced utterances can be longer than adults' in spite of significantly smaller lung volumes, due to differences in airflow through the glottis. On the other hand, children are predicted to produce fewer consonant-vowel syllables in an utterance because the durations of individual segments are usually longer. Formant frequencies and bandwidths are predicted using parameter values appropriate for child-sized vocal systems and models which are similar to those for adult speech. Values of frequency and bandwidth are not related to adult values by one simple scale factor. Measurements of young children's vocalizations are interpreted with reference to the models describing the relationships between acoustic characteristics and articulatory parameters. Developmental changes in vocal-tract characteristics, in respiratory and articulatory control, and in phonological representation are inferred from acoustic characteristics.
LANGUAGE-SPECIFIC PHONETIC PRODUCTION PATTERNS IN THE FIRST YEAR OF LIFE
Proceedings of the International Congress of Phonetic Sciences, 2007
The production of sounds by infants from 1 to 12 months is evaluated according to place of articulation to verify the hypothesis that infants' production becomes language-specific towards the end of the first year. This study is based on an analysis of 4,499 sounds produced by 19 infants raised in one of 3 linguistic contexts: Canadian English, Moroccan Arabic, and Bai. Our results reveal that towards the end of the first year (10-12 months), infants show a preference for producing sounds at places of articulation that reflect their linguistic background, a finding that parallels results obtained in perceptual studies. Contrary to our expectations, however, the infants' production at the end of the first year, albeit language-specific, does not directly correspond to the adult model. In the case of Bai infants, in particular, it was found that these infants use laryngeal constriction at a segmental level, while this phonetic feature is not employed in the adult model at a segmental level but at a syllable/word level for tone register contrasts.
A multi-modal, emergent view of the development of syllables in early phonology
Behavioral and Brain Sciences, 1998
A narrow focus on the jaw (or on motoric generators) does not account for individual and language-specific differences in babble and early speech. Furthermore, data from Christine Yoshinaga-Itano's laboratory at the University of Colorado support earlier findings that show glottal rather than oral stops in deaf infants' babbling:
Emergent complexity in early vocal acquisition: Cross linguistic comparisons of canonical babbling
Approaches to Phonological Complexity, 2009
Phonetic complexity, as evidenced in speech production patterns, is based on congruence of production system, perceptual, and cognitive capacities in adult speakers. Pre-linguistic vocalization patterns in human infants afford the opportunity to consider first stages in emergence of this complex system. The production system forms a primary site for considering determinates of early output complexity, as the respiratory, phonatory, and articulatory subsystems of infant humans support the types of vocal forms observed in early stages as well as those maintained in phonological systems of languages. The role of perceptual input from the environment in earliest stages of infant learning of ambient language phonological regularities is a second locus of emergent complexity. Young infants must both attend to and reproduce regularities to master the full range of phonological forms in their language. Cross linguistic comparison of babbling in infants acquiring typologically different languages including Dutch, Romanian, Turkish, Tunisian Arabic and French are described to consider production system based regularities and early perceptually based learning supporting emergence of ambient language phonological complexity.
Co-occurences of preverbal vocal behavior and motor action in early infancy
Developmental Science, 2001
This study reports on co-occurrence of vocal behaviors and motor actions in infants in the prelinguistic stage. Four Japanese infants were studied longitudinally from the age of 6 months to 11 months. For all the infants, a 40 min sample was coded for each monthly period. The vocalizations produced by the infants co-occurred with their rhythmic actions with high frequency, particularly in the period preceding the onset of canonical babbling. Acoustical analysis was conducted on the vocalizations recorded before and after the period when co-occurrence took place most frequently. Among the vocalizations recorded in the period when co-occurrence appeared most frequently, those that co-occurred with rhythmic action had significantly shorter syllable duration and shorter formant-frequency transition duration compared with those that did not co-occur with rhythmic action. The rapid transitions and short syllables were similar to patterns of duration found in mature speech. The acoustic features remained even after co-occurrence disappeared. These findings suggest that co-occurrence of rhythmic action and vocal behavior may contribute to the infant's acquisition of the ability to perform the rapid glottal and articulatory movements that are indispensable for spoken language acquisition.
Phonological development from babbling to speech: Common tendencies and individual differences
Applied Psycholinguistics, 1986
Taking as a point of departure Locke's biological model for the origins of phonological development, this study encompasses analyses of phonetic tendencies, consonant use in babbling and early words, and phonological word-selection patterns. Data from 10 children aged 9 to 16 months are drawn from four lexically defined points covering the period from no word use to a cumulative vocabulary of 50 words. Individual differences are found to prevail from the start in all three domains analyzed, with some increase in uniformity across subjects with increasing knowledge of language. Furthermore, the phonological processes typical of development from age 1 to 3 or 4 years are found to be rooted in the phonetic tendencies of the prelinguistic period.