Word Boundaries Research Articles

Lexical access across talkers

ABSTRACTThe current paper examined whether lexical access might (ever) cross talker boundaries. In four cross-modal priming experiments, listeners made visual lexical decisions after hearing an auditory word or word pair. On some trials, the auditory signal was produced by a single talker; on other trials, a talker gender change occurred in the middle of the auditory sequence. Participants demonstrated priming for items crossing both word boundaries and talker changes: after hearing a male talker say my and a female talker say great, participants showed a speeded response to “geese” (related to “migrate”). Priming was based on the combination of syllables across talkers, not driven simply by the first word. Findings suggest that acoustic cues indicating multiple talkers are insufficient to disrupt lexical access, and activation of lexical representations is not limited to those occurring within a single-talker stream. This provides support for models that do not involve explicit pre-lexical segmentation.

Probabilistic Phonotactics as a Cue for Recognizing Spoken Cantonese Words in Speech.

Previous experimental psycholinguistic studies suggested that the probabilistic phonotactics information might likely to hint the locations of word boundaries in continuous speech and hence posed an interesting solution to the empirical question on how we recognize/segment individual spoken word in speech. We investigated this issue by using Cantonese language as a testing case in the present study. A word-spotting task was used in which listeners were instructed to spot any Cantonese word from a series of nonsense sound sequences. We found that it was easier for the native Cantonese listeners to spot the target word in the nonsense sound sequences with high transitional probability phoneme combinations than those with low transitional probability phoneme combinations. These results concluded that native Cantonese listeners did make use of the transitional probability information to recognize the spoken word in speech.

How Speakers of Different Languages Extend Their Turns: Word Linking and Glottalization in French and German

ABSTRACTA speaker who issues a confirming turn starting with particles like yes, oui, ja, and so on, may mean to extend it and provide further material. This study shows that French and German speakers employ the same phonetic contrast to indicate the nature of that turn continuation. In spite of the typological difference between the German use of glottalization and the French use of linking phenomena for word boundaries involving word-initial vowels, speakers of both languages exploit this contrast systematically in their design of multiunit turns. Initial confirmations are joined directly to subsequent vowel-fronted turn components when speakers respond with an internally cohesive multiunit confirming turn. The components are separated by glottalization when responses involve multiple actions or departures from a trajectory projected by the turn-initial confirmation. This is further evidence that sound patterns shape interaction and are not solely determined by language-specific phonologies. Data are in French and German with English translation.

The role of continuous low-frequency harmonicity cues for interrupted speech perception in bimodal hearing.

Low-frequency acoustic cues have been shown to enhance speech perception by cochlear-implant users, particularly when target speech occurs in a competing background. The present study examined the extent to which a continuous representation of low-frequency harmonicity cues contributes to bimodal benefit in simulated bimodal listeners. Experiment 1 examined the benefit of restoring a continuous temporal envelope to the low-frequency ear while the vocoder ear received a temporally interrupted stimulus. Experiment 2 examined the effect of providing continuous harmonicity cues in the low-frequency ear as compared to restoring a continuous temporal envelope in the vocoder ear. Findings indicate that bimodal benefit for temporally interrupted speech increases when continuity is restored to either or both ears. The primary benefit appears to stem from the continuous temporal envelope in the low-frequency region providing additional phonetic cues related to manner and F1 frequency; a secondary contribution is provided by low-frequency harmonicity cues when a continuous representation of the temporal envelope is present in the low-frequency, or both ears. The continuous temporal envelope and harmonicity cues of low-frequency speech are thought to support bimodal benefit by facilitating identification of word and syllable boundaries, and by restoring partial phonetic cues that occur during gaps in the temporally interrupted stimulus.

Perception of Allophonic Cues to English Word Boundaries by Polish Learners: Approximant Devoicing in English

The study investigates the perception of devoicing of English /w, r, j, l/ after /p, t, k/ as a word-boundary cue by Polish listeners. Polish does not devoice sonorants following voiceless stops in word-initial positions. As a result, Polish learners are not made sensitive to sonorant devoicing as a segmentation cue. Higher-proficiency and lower-proficiency Polish learners of English participated in the task in which they recognised phrases such as buy train vs. bite rain or pie plot vs. pipe lot. The analysis of accuracy scores revealed that successful segmentation was only above chance level, indicating that sonorant voicing/devoicing cue was largely unattended to in identifying the boundary location. Moreover, higher proficiency did not lead to more successful segmentation. The analysis of reaction times showed an unclear pattern in which higher-proficiency listeners segmented the test phrases faster but not more accurately than lower-proficiency listeners. Finally, #CS sequences were recognised more accurately than C#S sequences, which was taken to suggest that the listeners may have had some limited knowledge that devoiced sonorants appear only in word-initial positions, but they treated voiced sonorants as equal candidates for word-final and word-initial positions.

Disentangling bilingualism from SLI: Dissociating exposure and input

Exposure and input are often used interchangeably in the study of child bilingualism to describe the learner's interaction with her environment. Carroll (Carroll), in her thought-provoking keynote article, challenges the equation between amount of exposure and amount of input necessary for acquiring a language, and in particular a second language. Carroll, whose diverse research borrows from both generative and psychological models of learning, brings these two perspectives into her exploration of the role of exposure and input in language acquisition, calling for a clearer definition of each. Carroll argues that the input to the language acquisition mechanism is the abstract construct necessary in order to solve the learning problem, and it is not a function of the amount of exposure: very little input is necessary in order to solve a learning problem, e.g., identifying word boundaries, or acquiring a syntactic structure. While scrutinizing the equation often drawn between exposure and input, Carroll further argues that the amount of exposure further misses the quality of this exposure as well as the societal context in which this exposure occurs.

Word boundaries in L2 speech: Evidence from Polish learners of English

Acoustic and perceptual studies investgate B2-level Polish learners’ acquisition of second language (L2) English word-boundaries involving word-initial vowels. In production, participants were less likely to produce glottalization of phrase-medial initial vowels in L2 English than in first language (L1) Polish. Perception studies employing word monitoring and word counting tasks found that glottalization of word-initial vowels had a negligible impact on the processing of L2 word boundaries. Taken together, these experiments suggest that B2-level learners are relatively successful in acquiring word-boundary linking processes that are for the most part absent from L1 Polish, and challenge the notion of an L2 ‘Word Integrity’ constraint. The cross-language interactions observed in these experiments are compatible with the claim that the realization of word-initial vowels is governed by a representational parameter, which may be derived in the Onset Prominence framework.

Resyllabification Reconsidered: On the Durational Properties of Word-Final /s/ in Spanish

Word-final consonants in Spanish are commonly assumed to undergo resyllabification across a word boundary before a following vowel, e.g., /los#otros/ 'the others' is realised as [lo.so.tros]. However, in many dialects of Spanish, word-final pre-vocalic consonants (‘derived onsets’) pattern phonologically with canonical codas and distinctly from canonical onsets. This property of derived onsets has been the subject of much interest in the phonological literature, and has led some linguists to question whether resyllabification indeed applies in all Spanish dialects. In this paper, we evaluate evidence for resyllabification based on acoustic data from 11 speakers of Peninsular Spanish. The results show that word-final pre-vocalic /s/ has increased duration compared to coda /s/, but at the same time, it is shorter compared to word-initial or word-medial pre-vocalic /s/. This result challenges an analysis where derived onsets become phonologically indistinguishable from canonical onsets. We consider an alternative in the form of partial resyllabification, and we further discuss the role of the syllable as a relevant unit in explaining /s/-sandhi in Spanish.

The Perception of Stress Pattern in Young Cochlear Implanted Children: An EEG Study.

Children with sensorineural hearing loss may (re)gain hearing with a cochlear implant—a device that transforms sounds into electric pulses and bypasses the dysfunctioning inner ear by stimulating the auditory nerve directly with an electrode array. Many implanted children master the acquisition of spoken language successfully, yet we still have little knowledge of the actual input they receive with the implant and specifically which language sensitive cues they hear. This would be important however, both for understanding the flexibility of the auditory system when presented with stimuli after a (life-) long phase of deprivation and for planning therapeutic intervention. In rhythmic languages the general stress pattern conveys important information about word boundaries. Infant language acquisition relies on such cues and can be severely hampered when this information is missing, as seen for dyslexic children and children with specific language impairment. Here we ask whether children with a cochlear implant perceive differences in stress patterns during their language acquisition phase and if they do, whether it is present directly following implant stimulation or if and how much time is needed for the auditory system to adapt to the new sensory modality. We performed a longitudinal ERP study, testing in bimonthly intervals the stress pattern perception of 17 young hearing impaired children (age range: 9–50 months; mean: 22 months) during their first 6 months of implant use. An additional session before the implantation served as control baseline. During a session they passively listened to an oddball paradigm featuring the disyllable “baba,” which was stressed either on the first or second syllable (trochaic vs. iambic stress pattern). A group of age-matched normal hearing children participated as controls. Our results show, that within the first 6 months of implant use the implanted children develop a negative mismatch response for iambic but not for trochaic deviants, thus showing the same result as the normal hearing controls. Even congenitally deaf children show the same developing pattern. We therefore conclude (a) that young implanted children have early access to stress pattern information and (b) that they develop ERP responses similar to those of normal hearing children.

The acoustic correlates of lexical stress in Israeli Hebrew

Lexical stress is an omnipresent phenomenon in spoken language, serving in many cases to disambiguate lexically identical words. Lexical stress in Spoken Israeli Hebrew is usually either penultimate or final, however its prosodic cues have not been measured systematically. In the present study the acoustics of lexical stress were characterized in detail. A list of 34 two-syllable stress-based minimal pairs was collected, where each word form has a different meaning. Each lexical word was embedded in a carrier sentence, thus creating 68 sentences, in which each word-form appeared once in sentence-middle position. Thirty speakers, gender balanced, uttered each sentence, giving a total of 2040 utterance productions. All syllable and vowel boundaries of the target words were annotated manually, and three acoustical parameters – duration, f0 and intensity – were measured for each nucleus vowel. Statistical analysis revealed that vowel duration was the dominant marker of Hebrew lexical stress; f0 played a minor role in indicating stress; while intensity played a more prominent role than f0. Regression analysis showed that the above three cues explained 80% of the variance, with duration contributing 77% and intensity and f0 1.5% each.

Differential Gaze Patterns on Eyes and Mouth During Audiovisual Speech Segmentation.

Speech is inextricably multisensory: both auditory and visual components provide critical information for all aspects of speech processing, including speech segmentation, the visual components of which have been the target of a growing number of studies. In particular, a recent study (Mitchel and Weiss, 2014) established that adults can utilize facial cues (i.e., visual prosody) to identify word boundaries in fluent speech. The current study expanded upon these results, using an eye tracker to identify highly attended facial features of the audiovisual display used in Mitchel and Weiss (2014). Subjects spent the most time watching the eyes and mouth. A significant trend in gaze durations was found with the longest gaze duration on the mouth, followed by the eyes and then the nose. In addition, eye-gaze patterns changed across familiarization as subjects learned the word boundaries, showing decreased attention to the mouth in later blocks while attention on other facial features remained consistent. These findings highlight the importance of the visual component of speech processing and suggest that the mouth may play a critical role in visual speech segmentation.

Sentiment detection in micro-blogs using unsupervised chunk extraction

In this paper, we present a proposed system designed for sentiment detection for micro-blog data in Chinese. Our system surprisingly benefits from the lack of word boundary in Chinese writing system and shifts the focus directly to larger and more relevant chunks. We use an unsupervised Chinese word segmentation system and binomial test to extract specific and endogenous lexicon chunks from the training corpus. We combine the lexicon chunks with other external resources to train a maximum entropy model for document classification. With this method, we obtained an averaged F1 score of 87.2 which outperforms the state-of-the-art approach based on the released data in the second SocialNLP shared task.

Sociophonetic study of young Nigerian English speakers

This study examines the variable use of r-liaison and boundary consonant deletion processes in the speech of young Nigerian speakers of English. This is with a view to confirming the hypothesis that continuous speech processes (CSPs) can be socially differentiated in a speech community. A sample of 180 young educated Nigerian English (NigE) speakers, evenly stratified into gender and class, voiced 19 utterances and a short passage into digital recording devices and filled in 180 copies of a structured questionnaire. All tokens of r-liaison and consonant deletion produced at word and morpheme boundaries were identified and analysed statistically, using the Analysis of Variance (ANOVA). The only speech variation observed in the data was between male and female speakers in boundary consonant deletion, (F(1, 176) = 6.24, p = .013). The findings did not sufficiently demonstrate variability in the speech patterns of young NigE speakers in relation to r-liaison and boundary consonant deletion processes. Keywords: connected speech, sociophonetics, r-liaison, boundary consonant deletion, Nigerian English

The informativeness of linguistic unit boundaries

Contemporary models of structural analysis tend to operate with discrete units at different linguistic levels. There is, however, considerable debate regarding the choice of units and the validity of the cues that guide their demarcation. At the level of grammatical analysis, this debate focuses largely on the status of words vs sub-word units and on the generality of the linguistic properties that mark each type of unit. This paper suggests that the status of a unit type can be evaluated in terms of its informativity A measure of informativity is obtained by assessing the influence that different unit boundary types have on text compressibility. The results obtained from this initial study support a pair of general conclusions. The first is that unit boundaries primarily reflect a statistical structure, and that the typological variability of linguistic cues reflects the fact that they serve a secondary reinforcing function. The second is that word boundaries are the most informative boundary type, and that the demarcation of words provides the most informative description of the regular patterns in a language.

How Transitional Probabilities and the Edge Effect Contribute to Listeners’ Phonological Bootstrapping Success

ABSTRACTMuch of what we know about the development of listeners’ word segmentation strategies originates from the artificial language-learning literature. However, many artificial speech streams designed to study word segmentation lack a salient cue found in all natural languages: utterance boundaries. In this study, participants listened to a speech-stream containing one of three sets of word boundary cues: transitional probabilities between syllables (TP Condition), silences marking utterance boundaries (UB Condition), or a combination of both cues (TP + UB Condition). Recognition of the trained words and rule words (words not in language, but conforming to its phonotactic structure) was tested. Participants performed equally well in the TP + UB and UB Conditions, scoring above chance on both trained and rule words. Performance in the TP condition, however, was at chance. Our results suggest that attention to UBs is a particularly effective strategy for finding words in speech, possibly providing a language-general solution to the word segmentation problem.

Preface

Preface Audrey Brunetaux (bio) and Michael S. Koppisch (bio) Seeing Charlotte Delbo / Seeing the Shoah Following World War II, survivors of the Holocaust struggled to recount an experience that remained beyond the boundary of words. How could Auschwitz be described in a way that someone who had not known its horrors might conceivably find factual? How to “make us see” the dehumanization and the extermination that were integral to what has come to be known as the univers concentrationnaire? A participant in the French Resistance and a Nazi concentration camp survivor, Charlotte Delbo has, especially because of the publication of her three-volume Auschwitz et après, been recognized as a major writer of Holocaust literature. Her unique style combines poetry with prose and employs a wide range of literary tools to convey the traumatic experience of the atrocities that she and her fellow prisoners endured daily. In her writings, Delbo challenges her readers to see her, to see her experience, to see the dreadful and horrifying mechanisms by which the camp universe functions. She incorporates into her narrative powerful images of the camps calculated, as she puts it, to “nous faire voir”—“make us see”—what happened in them. Painful details of an individual victim’s experience are noted on the page. Delbo’s originality lies in the conflation of the visual and the written to represent in her texts the unspeakable. In her highly visual writing, words become the eye of a camera, projecting onto the page a monstrous reality. These “shots,” made visible by the author’s rhetorical techniques, unveil the physical and psychological dimensions of the tragedy that might well remain obscured in strictly historical or journalistic accounts of events. Delbo’s traumatic, nonlinear memory calls up mental images that force the reader to confront the Holocaust as it destroyed individual lives. Reading Delbo is like viewing a film created by words that at once leave a trace on the page and become a repository in which is stored the memory of the dead. In a way, Delbo’s text becomes a site of memory not limited by the stasis of a monument. Although Delbo’s trilogy and others of her works were published by the Editions de Minuit, Delbo was not until recently seen as the major literary figure that she has become. Ironically, it was only after the English translation of Auschwitz et après by Yale University Press in 1995 that her work garnered the attention that it deserves. It is, in fact, still studied and written about more in English than in French. However, in the past few years, Delbo has been more widely read and studied in France, where the centennial of her birth was [End Page 7] celebrated in 2013 by an important conference at the Bibliothèque Nationale de France. There have thus far been but three collections of essays about Delbo’s work in French. The first, entitled Dossier Charlotte Delbo published in Témoigner entre histoire et mémoire: Revue pluridisciplinaire de la Fondation Auschwitz (2009) is a broad overview of Delbo’s style and literary engagement. More recently, David Caron and Sharon Marquart published Les Revenantes: Charlotte Delbo, la voix d’une communauté à jamais déportée (2011), essays that focus on the idea of community in Delbo’s writings. Another edited volume, Charlotte Delbo. Oeuvre et engagements (2014), included proceedings from the Charlotte Delbo Conference that took place in Paris in 2013 to celebrate the centennial of her birth. The English critic Nicole Thatcher has also written several monographs about Delbo—A Literary Analysis of Charlotte Delbo’s Concentration Camp Re-Presentation (1999) and Charlotte Delbo: Une Voix singulière (2003). Delbo’s work often figures centrally in studies related to French literature and the Shoah and has been the subject of scholarly pieces on trauma, poetry, drama, memory and literary representation. The author has become an important reference in Holocaust studies, and major critics in the field have written about Delbo’s poetic, minimalist and theatrical prose from innovative angles. For our project, we invited these scholars and others to contribute original pieces. Since no collection of its kind...

Icelandic post-lexical syllabification and vowel length in CVCV phonology

AbstractThe paper aims at providing a CVCV analysis of the Icelandic syllabification phenomena and the distribution of vocalic quantity. Two syllabification algorithms are reported to exist in Icelandic: lexical and post-lexical (Árnason 1998, 2011). The article will focus on the post-lexical algorithm, which determines vowel length in derivatives with class 2 suffixation, compounds, and across word boundaries. A modified model of CVCV will be argued for, which combines the insights from Scheer’s (2004, 2012) and Cyran’s (2003, 2010) model. It will be proposed that phonological computation applies only once to the whole string (preferably the whole sentence) and that the activity of the interface boils down to representational intervention (in accordance with Scheerian Direct Interface). It will be proposed that manipulation of Final Empty Nuclei (at least in the form of distributing parametric government) is a possible interface operation, as assumed in earlier CVCV (Scheer 2004). The complex pattern of two syllabification algorithms in Icelandic arises via an intricate interplay of sonority profiles of consonants and the needs of intervening empty nuclei (some of which are parametrically governed, while some others are not).

An overview of Kabarasi verb tone

This article presents the results of a study of verbal tone patterns in Kabarasi [lkb], a Kenyan Bantu language of the Luhya [luy] group. Kabarasi tone has a number of features that are common to Bantu languages (Kisseberth & Odden 2003, Downing 2011, Marlo & Odden 2017), including a lexical contrast between /H/ and /∅/ verb roots and a rich system of tonal inflection. Long H spans that extend across several words may be created by a pair of iterative, mutually feeding rules. One of these rules only applies across word boundaries and exhibits look-ahead effects; the other motivates a novel morphophonological domain: the limitative stem.

Tone assignment in Hong Kong English

This article provides an argument for Hong Kong English being a tonal language and informs the growing literature on word- and phrase-level prosody interactions. By teasing apart tonal effects that come from intonation and those that come from the word boundary, a clear picture emerges that H tones are assigned in all combinations to HKE di- and trisyllabic words. Tone spreading and blocking across words can also be seen in HKE, but syllables lexically specified for H never give up their tones. Complexity in HKE tone patterns arises when the H tones interact with boundary tones, such as the declarative final L% and the word-initial M.

Effects of interword spacing on native English readers of Chinese as a second language

AbstractInterword spacing facilitates English native readers but not native readers of Chinese, a writing system that does not mark word boundaries. L1-English readers of Chinese as a Second Language (CSL) could then be facilitated if spacing is added between words in Chinese materials. However, previous studies produced inconsistent results. This study tested the hypothesis that interword spacing facilitates L1-English CSL readers. We used an online multiple-choice gap-filling task to test 12 English CSL readers and 12 Chinese natives reading a series of eight texts of suitable difficulty, written with or without interword spacing. The CSL readers read faster with interword spacing than without, while Chinese native readers were not affected. The interword spacing effect was negatively correlated with measures of reading proficiency. It is argued that interword spacing facilitates CSL readers reading materials of sufficient complexity by facilitating their lexical parsing. Pedagogical implications are discussed.