Search

Highlights • 61 clusters in 174 words were syllabified by 30 Czech speakers in two experiments. • Sonority type exercised a significant influence, following sonority sequencing. • Vowel length effects were found not induced by phonotactic constraints. • Onset frequency of occurrence and morphology also affected the results. • Task-related differences in syllabification patterns were identified. Abstract This study examines syllabification preferences of 30 speakers of Czech in two behavioural experiments using real disyllabic words with 61 intervocalic CC clusters as stimuli. The aim was to evaluate competing theoretical predictions about syllable boundaries in Czech. Participants synchronized individual syllables with metronome pulses in Experiment 1 (induced pause insertion) and produced syllables in reversed order in Experiment 2 (syllable reversal). Logistic regression analyses revealed significant effects of cluster sonority type, phonological length of the preceding vowel and word-edge phonotactics (also in relation to frequency of occurrence). Morphological structure of the items significantly influenced syllable boundary placement as well. The results of both experiments converge towards the effects found in previous studies on English and some other languages. However, ambisyllabic responses were virtually non-existent in pause insertion and relatively low (8%) in syllable reversal, which differs from the results on Germanic languages. Finally, the findings do not support strict onset maximization but rather indicate an onset-filling strategy. [ABSTRACT FROM AUTHOR]

The present paper explores the dynamics of speech production and perception in the context of syllabification and categorization. The selective review includes empirical work and dynamical models that account for changes in the perception and production of syllable structure as transitions between attractors in a dynamical system and that highlight the role of instabilities as a mechanism for regulating flexibility and change. Different conceptual approaches to changes in perceptual categorization are reviewed, including a nonlinear dynamic model, a related Bayesian approach, and a hybrid approach. Of particular importance are recent models that incorporate cognitive factors (such as attention, expectation, and memory) and that change slowly or quickly relative to the changing acoustic input. These dynamical models allow phenomena such as self-organization, emergence, and other hallmarks of complex adaptive systems and may also suggest a mechanism linking speech production and perception, providing an alternative description to the internal models often invoked. [ABSTRACT FROM AUTHOR]

This paper investigates how strong the influences of syllables are on the organization of speech movements in Mandarin Chinese and French. More specifically, we evaluate the strength of these influences by measuring the extent of anticipatory coarticulation in V 1 CV 2 sequences, within and across the boundaries of the CV 2 syllable. In line with motor control studies that have investigated serial-order motor tasks such as speech production, we consider that the extent of anticipatory motor planning reveals information about the units of planning, their boundaries and their strengths. For V 1 /t/V 2 and V 1 /k/V 2 sequences, where V 1 and V 2 are either /i/, /a/ or /u/, we used an electromagnetic magnetometer to record articulatory data from three native speakers of Mandarin Chinese and three native speakers of French. We labeled key articulatory configurations for V 1 , C and V 2 based on general articulatory criteria. Our two major findings are: (1) Within CV 2 boundaries, coarticulation is strong for both groups of speakers; in French, the coarticulation patterns can be fully explained by anticipatory behavior, while in Mandarin Chinese other strategies might also be involved. (2) Across CV 2 boundaries, anticipatory coarticulation is observed in numerous cases in French, while no significant anticipation can be noted in Mandarin Chinese. We conclude that these measures provide evidence for the hypothesis that the strength of the syllable as a unit of speech planning is more important in Mandarin Chinese than in French. Finally, we propose potential explanations for these differences. [ABSTRACT FROM AUTHOR]

Most computational models that have addressed the development of consonant–vowel syllable systems have assumed a preexisting tendency to produce syllabically structured utterances. For instance, the COSMO model assumes a consistent base of jaw movement upon which finer-grained articulatory patterns are learned, motivated by MacNeilage and Davis's frame-then-content theory (FCT). While research operating under this assumption has provided much useful information on infant speech development, it does not address the gradual transition from non-syllabic to syllabic sounds that occurs during the first seven months of human infancy. It is important to consider the role that learning plays in this very significant transition. This paper discusses two computational models that address how infants may learn to produce syllabic utterances. Future work should develop agent-based models of sound production that show how syllabicity itself can emerge in a population of agents. [ABSTRACT FROM AUTHOR]

Elevated fundamental frequency ( F 0) has been found to have similar properties across languages. For example, raised pitch often accompanies syllabic stress, emotionally charged speech, infant-directed speech, and questions. In many languages, occurrence of high tone is subject to more constraints than are other tones. Given that these patterns occur frequently in the world's languages, it is natural to ask whether language-independent properties of raised F 0 could play a role in the existence of typological similarities. The purpose of the present paper is to survey possible language-external factors that appear to play a role in the special status of linguistic H(igh). Moreover, the collection of studies assembled in this Special Issue provides empirical evidence that raised F 0 attracts listener attention differently from lowered F 0, that sustained production of high F 0 may involve unique auditory control mechanisms, and that social context and even semantics may contribute to speaker production of raised F 0. It is hoped that the articles of the special issue will provide a phonetic basis to explain some of the asymmetries observable in prosodic systems of languages around the world. [ABSTRACT FROM AUTHOR]

Abstract: The durational realization of the Swedish complementary quantity contrast was investigated in monosyllabic and disyllabic target words in Central Standard Swedish and in Fenno-Swedish. Several systematic durational differences between the two varieties were observed, most of which can be summarized by stating that speakers of Fenno-Swedish exaggerate, in comparison to Central Standard Swedish speakers, the quantity contrast between the vowel in the stressed syllable and the following consonant, i.e. they make short segments shorter and long segments longer. This we explain as influence on Fenno-Swedish from Finnish, in which vowels and consonants have a binary quantity contrast independent of each other. In speaking and listening to Finnish, bilingual speakers of Fenno-Swedish have come to make larger durational distinctions than occur in Central Standard Swedish. [Copyright &y& Elsevier]

Abstract: This paper reports the results of an English experiment on vowel-shortening in different contexts. The data concern compression effects, whereby, in syllables with a greater number of segments, each one of the segments is shorter than in syllables with fewer segments. The experiment demonstrates that the amount of vowel compression found in English monosyllabic words depends in part on which consonants occur adjacent to the vowel in that word, how many consonants occur, and in which position they occur. Consonant clusters drive more vowel shortening than singletons when they involve liquids, but not when they involve only obstruents. Clusters involving nasals drive shortening relative to singletons only in onset position. We suggest that the results cannot be reduced to general principles of gestural overlap and coordination between consonants and vowels, but instead require a theory with overt representation of auditory duration. [Copyright &y& Elsevier]

Abstract: A pre-vocalic connected speech context is said to enable the resyllabification of word-final consonants into an onset, thus conditioning alternations. We present EPG data on English word-final /l/, measuring the extent of alveolar contact and the rate of vocalisation, the extent of dorsal retraction (representing “darkness”), and the timing of alveolar contact relative to dorsal retraction. Two dialects of British English are considered, namely Scottish Standard English and Southern Standard British English. Results are that /l/ alternation is systematic: the tongue tip contact of word-final /l/, quite categorically for some speakers, is more onset-like in pre-vocalic and more coda-like in pre-consonantal contexts. This alternation is not along the lines predicted by a segmental resyllabification account, however. First, the segmental identity of the following consonant (/b/ or /h/) may be as powerful a factor in conditioning the presence or absence of alveolar contact for some speakers. Second, glottalisation of lexically vowel-initial words regularly occurs, but does not seem to condition the appearance (or otherwise) of tongue tip contact. Third, the tongue dorsum remains retracted and does not adopt an onset-like form or timing even when /l/ is pre-vocalic. Thus categorical resyllabification of a word-final /l/ segment based on phonotactic acceptability is rejected as a mechanism controlling English L-sandhi in connected speech. Instead, we propose a gestural-episodic model, in which individual gestures display different levels of coherence in lexical syllable roles, while in connected speech, segmental sequences are influenced by similarity to well-rehearsed lexical sequences, if they exist. [Copyright &y& Elsevier]

Abstract: We investigated the perceptibility of stop voicing in a domain-final neutralizing context in German that according to various phonological models is completely neutralized in favour of the voiceless category but that according to various empirical studies is distinguishable phonetically. A primary aim was to determine whether acoustic cues that were available for the stop voicing distinction were perceptible in a neutralizing context. A secondary aim was to assess whether voicing perception was influenced by phonotactic frequency and the potential for resyllabification. Nineteen listeners of a Standard German speaking variety made forced-choice judgments to synthetic stimuli spanning a voiced–voiceless continuum containing domain-final alveolar and velar stops in various neutralizing contexts that differed in terms of phonotactic probability and the potential for resyllabification. Our results showed that voicing information could be distinguished but that the perceptibility of this distinction also depended on statistical properties of phoneme sequences and whether a domain-final stop could potentially be perceptually resyllabified as domain-initial. Our general conclusion is that a categorical neutralization model is insufficient to account for stop voicing perception in German in a domain-final context: instead, voicing perceptibility in these contexts depends on an interaction between acoustic information and phonological knowledge which emerges as a generalization across the lexicon. [Copyright &y& Elsevier]

Abstract: This study examines how native language (L1) experience affects auditory–visual (AV) perception of nonnative (L2) speech. Korean, Mandarin and English perceivers were presented with English CV syllables containing fricatives with three places of articulation: labiodentals nonexistent in Korean, interdentals nonexistent in Korean and Mandarin, and alveolars occurring in all three L1s. The stimuli were presented as auditory-only, visual-only, congruent AV and incongruent AV. Results show that for the labiodentals which are nonnative in Korean, the Koreans had lower accuracy for the visual domain than the English and the Mandarin perceivers, but they nevertheless achieved native-level perception in the auditory and AV domains. For the interdentals nonexistent in Korean and Mandarin, while both nonnative groups had lower accuracy in the auditory domain than the native English group, they benefited from the visual information with improved performance in AV perception. Comparing the two nonnative groups, the Mandarin perceivers showed poorer auditory and AV identification for the interdentals and greater AV-fusion with the incongruent AV material than did the Koreans. These results indicate that nonnative perceivers are able to use visual speech information in L2 perception, although acquiring accurate use of the auditory and visual domains may not be similarly achieved across native groups, a process influenced by L1 experience. [Copyright &y& Elsevier]

Abstract: Listener syllabification judgments vary on words with VCCV sequences. This variability seems to present a challenge to phonological theory, which predicts invariant behavior following from rules or principles. Judgment variability might be better accounted for by positing that syllable production and perception are linked by the signal. According to such a hypothesis, variable listener judgments would result from variable speaker productions. To determine whether phonological or phonetic factors better account for listener syllabification judgments, different speakers produced nonsense words, listeners syllabified them, and then stimuli characteristics were used to predict syllabification. Results showed that, in spite of production variability, listener judgments were nearly invariant on words with medial sequences that formed illegal onset clusters to the second syllable and on words with first syllable stress, suggesting that these tokens were syllabified according to categorical phonological knowledge. Other judgments could not be similarly explained, but instead were best predicted by gradient phonetic patterns cuing juncture. The results disconfirm the hypothesis that syllable production and perception are directly linked via the signal. Instead, they suggest a two-step model of English syllabification in which listeners rely on juncture cues to determine syllable boundaries only after phonological knowledge fails to indicate a unique boundary location. [Copyright &y& Elsevier]