Whole-word Repetitions Research Articles

Enhancing consecutive interpreting with ASR: Sight-Terp as a computer-assisted interpreting tool

This experimental study investigates the potential impact of employing automatic speech recognition (ASR) and speech translation (ST) in consecutive interpreting (CI) through the use of a computer-assisted interpreting (CAI) tool. The tool used is Sight-Terp, an ASR-supported CAI tool developed and designed by the first author of this study. It offers multiple features, such as ASR, real-time ST, named entity highlighting, and automatically enumerated segmentation. The methodology adopted in this research involves a within-subjects design, assessing participants’ output in scenarios with and without the use of Sight-Terp on a tablet. Twelve participants were recruited for the experimental setup and asked to interpret four English speeches into Turkish in long CI mode, using Sight-Terp for two of them and a pen and paper for the other two. The data analysis is grounded on parameters of both accuracy and fluency. In distinguishing the variance in accuracy across the two settings, accuracy metrics were rooted in the mean count of correctly rendered semantic units (units of meaning), as defined by Seleskovitch (1989). On the other hand, fluency was quantified by tracking the frequency of disfluency markers, including false starts, frequency of filled pauses, filler words, whole-word repetitions, broken words, and incomplete phrases in each session. The results show that the integration of ASR into two CI tasks led to an enhancement in the accuracy of the participants’ rendition. Concurrently, however, it led to an increase in disfluencies and extended task durations compared to the tasks in which Sight-Terp was not used. The study outcomes also suggest potential areas of improvement and modifications that could further enhance the utility of the tool. Future empirical studies using Sight-Terp will tell us more about the feasibility of ASR in the interpreting process and cognitive aspects of human-machine interaction in CI.

Comparison of performance of automatic recognizers for stutters in speech trained with event or interval markers.

Automatic recognition of stutters (ARS) from speech recordings can facilitate objective assessment and intervention for people who stutter. However, the performance of ARS systems may depend on how the speech data are segmented and labelled for training and testing. This study compared two segmentation methods: event-based, which delimits speech segments by their fluency status, and interval-based, which uses fixed-length segments regardless of fluency. Machine learning models were trained and evaluated on interval-based and event-based stuttered speech corpora. The models used acoustic and linguistic features extracted from the speech signal and the transcriptions generated by a state-of-the-art automatic speech recognition system. The results showed that event-based segmentation led to better ARS performance than interval-based segmentation, as measured by the area under the curve (AUC) of the receiver operating characteristic. The results suggest differences in the quality and quantity of the data because of segmentation method. The inclusion of linguistic features improved the detection of whole-word repetitions, but not other types of stutters. The findings suggest that event-based segmentation is more suitable for ARS than interval-based segmentation, as it preserves the exact boundaries and types of stutters. The linguistic features provide useful information for separating supra-lexical disfluencies from fluent speech but may not capture the acoustic characteristics of stutters. Future work should explore more robust and diverse features, as well as larger and more representative datasets, for developing effective ARS systems.

A Comparison of Speech Dysfluencies Before and After Sports-Related Concussion

<h3>Research Objectives</h3> The purpose of this study is to compare the number and type of speech disfluencies that occur in college-aged athletes at baseline compared to after a sports-related concussion. <h3>Design</h3> Pre-post (repeated measures). <h3>Setting</h3> University. <h3>Participants</h3> A total of 230 Division I student athletes participated in pre-season speech testing. Twelve students (18-22 years) who sustained a concussion participated in speech testing in the days following concussion. Picture description speech samples were independently coded by three trained raters for 17 error types in the traditional categories of errors defined in fluency analysis. <h3>Interventions</h3> Not Applicable. <h3>Main Outcome Measures</h3> Error types were sorted into three categories: stuttering-like dysfluencies (part-word repetition, single-syllable whole-word repetition, prolongation, block), articulation errors (substitution, distortion, addition, omission), and other dysfluencies (multisyllable whole-word repetition, revision, abandoned or incomplete segment, interjection, filler, phrase repetition, pauses). <h3>Results</h3> Within-subjects analysis comparing the difference in percent error scores at baseline and post-concussion revealed significant differences for interjections (t(11)=-2.678, p< .05). The Other Disfluency category was also significantly different (t(11)= -2.735, p< .05), with more errors occurring after a concussion. No change in the Stuttering-Like Disfluency (t(11)= -0.799, p>.05) or Articulation Error category (t(11)=-0.045, p>.05) was found. <h3>Conclusions</h3> As predicted, pre-post comparisons demonstrated an overall increase in total number of speech errors after a concussion (Mean Baseline = 11.5 errors; Mean Concussion = 15.9 errors). Results demonstrate that specific speech fluency changes occur following mild sports-related concussions. Future studies should consider speech as a diagnostic tool for concussion. <h3>Author(s) Disclosures</h3> Funding for this work is through an internal grant by Seton Hall University. No other relevant disclosures.

The effect of cognitive load on temporal and disfluency patterns of speech

AbstractThis paper investigates how variation in the complexity of speech tasks is reflected in the temporal characteristics and disfluency patterns of speech. We examined temporal characteristics (speech rate, global articulation rate, ratio of pauses, frequency of pauses, and mean duration of pauses) and disfluency markers (overall frequency of disfluencies; frequency of filled pauses, filler words, whole-word repetitions, part-word repetitions, broken words, prolonged sounds, and revisions; frequency of disfluency clusters) in four speech production tasks (consecutive interpreting, sight translation, spontaneous speech and extemporaneous speech) with twelve speakers.Our hypothesis, according to which the examined parameters will differ across the four tasks, was partly confirmed by the data; even though not all speech tasks differed significantly in all the examined parameters, our investigation revealed that there were significant differences between some tasks in four parameters, and between others in nine out of the fourteen parameters examined. Our data also suggest that in terms of the temporal characteristics and disfluency markers examined, the four tasks can be represented on a continuum based on the cognitive load associated with each task. At one end of the continuum and generating the least cognitive load is spontaneous speech, and at the other, generating the most cognitive load, is sight translation.

DISFLUENCIES IN STAND-UP COMEDY: A PSYCHOLINGUISTIC ANALYSIS ON DREW LYNCH'S STUTTERING

Difficulties of producing speech sound in stutterers are indicated by the repetition, pause, prolongation, revision, and filled pause on the speaking. However, such difficulties do not hinder the communication as shown in the speech of a stand-up comedian named Drew Lynch. This study aims at exploring the types of fluency disorder identified in Lynch’s utterances on stage. This study uses the descriptive qualitative method employed through the process of observing, transcribing, describing, and analyzing his utterances in American Got Talent videos. The result shows Lynch produces all kinds of disfluency covering filled pause, phrase repetition, revision, multisyllabic whole-word repetition, monosyllabic whole-word repetition, repetition of individual sound or syllable, prolongation of sound, and block. The monosyllabic whole-word repetition is more dominant. The combination happens between revision with monosyllabic whole-word repetition, prolongation, or multisyllabic whole-word repetition. These findings confirm that in the context of stand-up comedy, the disfluencies in stuttering do not hamper the transfer of meaning.

Phonological neighborhood effect in spontaneous speech in adults who stutter

PurposeThe study examined the relationship between word-form similarity network (phonological neighborhood) and stuttering occurrence in spontaneous speech in adults. The study asked whether neighborhood characteristics, including the number of neighbors (neighborhood density) and the average word frequency among neighbors (neighborhood frequency), differentiate stuttered from fluent words within spontaneous speech samples, and more specifically, whether neighborhood characteristics facilitate speech fluency in adults who stutter. MethodSpontaneous speech samples were collected from 14 adults who stutter, including 7 with mild stuttering and 7 with severe stuttering. Each stuttered word was matched with a fluently produced word within the sample, controlling for multiple lexical factors. Neighborhood density and neighborhood frequency were compared between the stuttered and matched fluent words. ResultsAdults stuttered on words with lower neighborhood density, compared to fluently produced words, with similar patterns in the two severity subgroups. There appeared to be a marginal difference between single-syllable whole-word repetition and the other stuttering typologies. ConclusionNeighborhood density facilitates fluent word production in spontaneous speech in adults who stutter, regardless of stuttering severity. This finding adds to the evidence supporting that phonological encoding plays a role in stuttering in naturalistic, spontaneous speech in adults.

Disfluent whole-word repetitions in cluttering: Durational patterns and functions

ABSTRACTIn this article, disfluent word-repetitions are analysed in cluttered and control speech. The main questions are the following: (1) Do different functions of word-repetitions occur in different ratio in cluttered and control speech? (2) Are there any differences between PWC and control speakers in durational parameters of disfluent word-repetitions?Results show that there are differences between the two groups of speakers in the types of word-repetitions, and in their durational parameters. In cluttered speech, the most frequent type of repetitions were covert self-repairs while in control speech canonical repetitions are dominated. There were significant differences in the durational parameters of word-repetitions of both groups except in the function of covert self-monitoring.According to the results, it can be concluded that as in earlier studies, analysis of the frequency of types of disfluencies in themselves is not enough as this analysis does not always show the differences between the two groups. It is important to learn more about the durational and functional characteristics of the specific types of disfluencies comparing them in cluttered and control speech.

A Case of Afferent Apraxic Motor Aphasia? A Tribute to A.R.Luria

A Case of Afferent Apraxic Motor Aphasia? A Tribute to A.R.Luria

An Investigation into the Difference in Word Class and the Function of Repetition in Stuttering Persian Speaking Children

Various studies have explored developmental stuttering in children. However, not many investigations have been conducted to examine repetition types produced by these children. That said, the purpose of this study was to examine repetition types produced by the children who stutter and those who do not, observing for differences in word classes, including content and function words. To that end, six children who stutter and eight children who do not participated in this study. The participants were Iranian and they were monolingual Persian speakers. The language productions of these children were recorded. A situation has been created for them in which they produced natural data. The results of the study indicated that both groups of children produced more number of partword repetition than whole-word repetition. Moreover, the results indicated that there was a statistically significant difference between the two groups with regard to the repetition types of different word classes. The findings also revealed that children who stutter produced significantly more part-word repetition in both content and function words. This was also observed for the content words produced by the children who do not stutter. However, no significant difference was observed in the repetition types of function words in the production of the children who do not stutter. The study has proposed some arguments for the obtained results.

Stimulus-driven changes in the direction of neural priming during visual word recognition

Visual object recognition is generally known to be facilitated when targets are preceded by the same or relevant stimuli. For written words, however, the beneficial effect of priming can be reversed when primes and targets share initial syllables (e.g., “boca” and “bono”). Using fMRI, the present study explored neuroanatomical correlates of this negative syllabic priming. In each trial, participants made semantic judgment about a centrally presented target, which was preceded by a masked prime flashed either to the left or right visual field. We observed that the inhibitory priming during reading was associated with a left-lateralized effect of repetition enhancement in the inferior frontal gyrus (IFG), rather than repetition suppression in the ventral visual region previously associated with facilitatory behavioral priming. We further performed a second fMRI experiment using a classical whole-word repetition priming paradigm with the same hemifield procedure and task instruction, and obtained well-known effects of repetition suppression in the left occipito-temporal cortex. These results therefore suggest that the left IFG constitutes a fast word processing system distinct from the posterior visual word-form system and that the directions of repetition effects can change with intrinsic properties of stimuli even when participants' cognitive and attentional states are kept constant.

Bupropion-Induced Stuttering.

To the Editor: Stuttering is defined as frequent occurrences of sound and syllable repetitions, sound prolongations, interjections, broken words, audible or silent blocking, circumlocutions, words produced with an excess of physical tension, and monosyllabic whole-word repetitions.1 Stuttering should be classified as a disorder only if its severity is such as to markedly disturb the fluency of speech. Stuttering is usually developmental, but, in rare cases, it may be acquired in adulthood as the result of a neurologic event such as a head injury, tumor, stroke, or drug use.1 Some drugs that have been reported to cause stuttering are antipsychotics, antidepressants, mood stabilizers, and others.2,3 Bupropion is an effective antidepressant, and stuttering due to the drug has not been reported.4 There is only 1 previously reported case of stuttering due to bupropion5 in which the patient remitted after discontinuation of the drug, but no rechallenge was given. In the present case, stuttering disappeared after discontinuation of bupropion and reappeared on rechallenge. Case report. Mr A, a 38-year-old married man, presented to the hospital with complaints of sad mood, loss of interest in surroundings, hopelessness and worthlessness, lethargy, heaviness in head, decreased appetite and sleep, loss of libido, and crying spells increasing gradually for the last 6 weeks. There was no past or family history of a psychiatric disorder. He was a college graduate and had a history of occasional smoking. The mental status examination revealed psychomotor retardation and sad mood. Speech was normal in rate and volume with increased reaction time. There were ideas of hopelessness and worthlessness. Mr A was diagnosed with major depressive disorder (DSM-IV criteria). The laboratory investigations including routine hematology, blood sugar, liver, and renal and thyroid function tests as well as computed tomography scan (head) were normal. Mr A was started on tablet bupropion slow release 150 mg daily. After 2 weeks, he reported mild improvement. The dose was increased to 300 mg daily. After 2 days of the increased dose, he reported decreased, labored, and pressured speech. There were involuntary silent pauses or blocks and repetitions and prolongations of sounds, syllables, and words. Rhythm of speech was also affected. Mr A’s wife reported that the increased dose of bupropion had resulted in his speech problem, which he had not experienced previously. There was no disorientation or unconsciousness during stuttering. Bupropion was stopped immediately, and there was instant improvement in his speech problem. Normal rhythm of speech was restored and intensity normalized. Rechallenge with bupropion 150 mg after 1 week caused stammering and was stopped immediately. The role of several neurotransmitters such as γ-aminobutyric acid, serotonin, and dopamine has been proposed in the pathogenesis of stuttering. Bupropion’s ability to increase dopamine in the frontal cortex was suspected to have been involved in this patient’s stuttering.5

The function of repeating: The relation between word class and repetition type in developmental stuttering.

It is already known that preschool-age children who stutter (CWS) tend to stutter on function words at the beginning of sentences. It is also known that phonological errors potentially resulting in part-word repetitions tend to occur on content words. However, the precise relation between word class and repetition type in preschool-age stuttering is unknown. To investigate repetitions associated with monosyllabic words in preschool-age CWS. Specifically, it was hypothesized that repetition type should vary according to word class in preschool-age CWS and children who do not stutter (CWNS). Thirteen preschool-age CWS and 15 preschool-age CWNS produced age-appropriate narratives, which were transcribed and coded for part-word repetitions (PWR) and whole-word repetitions (WWR) occurring on monosyllabic words. Each repetition type was also coded for word class (i.e., function versus content). Results indicated that although CWS and CWNS were significantly more likely to produce PWR on content words, this tendency did not differ between the two talker groups. Further, CWS and CWNS did not differ in their tendencies to produce PWR versus WWR overall, but the tendency to produce repetitions on function words was significantly greater for CWS versus CWNS. Findings are taken to suggest that repetitions of monosyllabic words in young children are not easily explained from the perspective of phonological errors, but may instead be considered from an incremental planning of speech perspective.

Automatic Recognition of Repetitions in Stuttered Speech: Using End-Point Detection and Dynamic Time Warping

This study proposes a methodology for recognizing repetitions in stuttered speech. First, the recorded speech is parameterized by extracting six acoustic features, including volume, zero crossing rate, spectral entropy, high-order derivatives, VH curve, and VE curve. Second, the speech is segmented using the technique of end-point detection according (EPD) to the threshold of VH curve. Third, the features of the segmented speech are processed by dynamic time warping (DTW) to identify similar patterns in neighbouring segments. The proposed method was verified using the artificial stuttering samples of Mandarin Chinese. Ten male subjects were asked to imitate stuttering by speak out 39 predefined repetition settings. These settings are planned by considering three Mandarin Phonetic Symbols ([t], [k], [t‘]) and three kinds of repetitions (part-word repetition, whole-word repetition, multi-syllable word repetition). The experimental results indicate that EPD using VH curve is capable to slice the repetition in artificial stuttered speech. Comparing the results for recognizing the phoneme and single syllable words, there is no significant difference for the threshold of DTW. The performance of DTW in recognizing repetitions had high accuracy of 83%. Therefore, the proposed method combining EPD and DTW is feasible for automatic recognition of repetitions in stuttered speech. However, more real stuttered speech samples are still needed to verify and improve the proposed method.

Lifestyle profile of school-aged children suffering from pathological stuttering

Abstract Stuttering is a speech disorder that involves intraphonemic disruption, part-word repetitions, monosyllabic whole-word repetitions, prolongation, and silent fixations (blocks). The aim of the study was to investigate the lifestyle profile of children suffering from pathological stuttering and identify the factors that worsen or improve the condition of the child with pathological stuttering. The study sample consisted of 60 children suffering from pathological stuttering as well as their mothers or caregivers who attended the previous setting. Data were collected using two tools. The first tool was a structured questionnaire that included biosocial characteristics of the children and biosocial data of mothers. The second tool was an observation checklist developed by the researcher to observe children suffering from pathological stuttering as well as their mothers during speech therapy. It was found that 63.3% of children with stuttering were of a mean age of 8.17 ± 1.66 years. Significant difference was found between the mean age of stuttering children and their socialization skills. Significant difference was found between the mean age of stuttering children and response to treatment. From the present study it can be concluded that pathological stuttering as a disease is easy to diagnose, difficult to treat, and has many negative impacts on the physical, psychological, social, and spiritual aspect of the child.

Lexical category influences in Persian children who stutter

This article explores the effect that words from different lexical categories have on disfluency in 12 Persian children, ten boys and two girls, who stutter. They were aged 7 years 5 months to 10 years 6 months. Words from the participants’ narrative and reading samples (sub-tests of the Reading and Dyslexia Test validated for Persian school-aged children) were categorized as content, function, or content–function, and stuttering-like disfluencies were coded in each speech sample. Content and content–function words were significantly more likely to show stuttering-like disfluencies than function words. The distribution of symptom types over content and content–function words was similar, and differed from the distribution seen in function words. The symptom type analysis also supported the view that whole-word repetitions should not be grouped with other stuttering-like disfluencies.

Screening school-aged children for risk of stuttering

ObjectivesHowell and Davis's (2011) model that predicts whether stuttering in eight-year old children will persist or recover by teenage was adapted for screening school-aged children for risk of stuttering. Stuttering-severity scores were used to predict whether children belonged to fluent or stuttering groups. Predicted group assignments were compared for models in which severity measures were made with whole-word repetitions excluded or included. The best model for distinguishing children who stutter (CWS) from fluent children was validated across a wide range of ages. DesignStuttering-severity scores from CWS (222 for development, and 272 for validation, of the models) and fluent children (103 for development, and 25 for validation, of the models) were employed. Models were developed that predicted prognosis and screened CWS and fluent children. All these analyses were conducted both with whole-word repetitions excluded and included in the stuttering-severity scores. The model that screened fluent children from all CWS which excluded whole-word repetitions was validated for children across a range of ages. ResultsAll models achieved around 80% specificity and sensitivity. Models in which whole-word repetitions were excluded were always better than those which included them. Validation of the screening for fluency with whole-word repetitions excluded classified 84.4% of fluent children, and 88.0% of CWS, correctly. Some of these children differed in age from those used to develop the model. ConclusionHowell and Davis's risk factor model for predicting persistence/recovery can be extended to screen school-aged children for fluency.Educational objectives: After reading this article, participants will be able to: (1) describe the difference between finding group differences and risk factor modeling in stuttering research; (2) summarize the strengths and weaknesses of stuttering severity instrument version three; (3) discuss how validation of diagnostic and screening models for fluency can be performed; (4) see how risk models have potential applications for screening for communication disorders in general.

Screening school-aged children for risk of stuttering

ObjectivesHowell and Davis's (2011) model that predicts whether stuttering in eight-year old children will persist or recover by teenage was adapted for screening school-aged children for risk of stuttering. Stuttering-severity scores were used to predict whether children belonged to fluent or stuttering groups. Predicted group assignments were compared for models in which severity measures were made with whole-word repetitions excluded or included. The best model for distinguishing children who stutter (CWS) from fluent children was validated across a wide range of ages. DesignStuttering-severity scores from CWS (222 for development, and 272 for validation, of the models) and fluent children (103 for development, and 25 for validation, of the models) were employed. Models were developed that predicted prognosis and screened CWS and fluent children. All these analyses were conducted both with whole-word repetitions excluded and included in the stuttering-severity scores. The model that screened fluent children from all CWS which excluded whole-word repetitions was validated for children across a range of ages. ResultsAll models achieved around 80% specificity and sensitivity. Models in which whole-word repetitions were excluded were always better than those which included them. Validation of the screening for fluency with whole-word repetitions excluded classified 84.4% of fluent children, and 88.0% of CWS, correctly. Some of these children differed in age from those used to develop the model. ConclusionHowell and Davis's risk factor model for predicting persistence/recovery can be extended to screen school-aged children for fluency.

Classification of types of stuttering symptoms based on brain activity.

Among the non-fluencies seen in speech, some are more typical (MT) of stuttering speakers, whereas others are less typical (LT) and are common to both stuttering and fluent speakers. No neuroimaging work has evaluated the neural basis for grouping these symptom types. Another long-debated issue is which type (LT, MT) whole-word repetitions (WWR) should be placed in. In this study, a sentence completion task was performed by twenty stuttering patients who were scanned using an event-related design. This task elicited stuttering in these patients. Each stuttered trial from each patient was sorted into the MT or LT types with WWR put aside. Pattern classification was employed to train a patient-specific single trial model to automatically classify each trial as MT or LT using the corresponding fMRI data. This model was then validated by using test data that were independent of the training data. In a subsequent analysis, the classification model, just established, was used to determine which type the WWR should be placed in. The results showed that the LT and the MT could be separated with high accuracy based on their brain activity. The brain regions that made most contribution to the separation of the types were: the left inferior frontal cortex and bilateral precuneus, both of which showed higher activity in the MT than in the LT; and the left putamen and right cerebellum which showed the opposite activity pattern. The results also showed that the brain activity for WWR was more similar to that of the LT and fluent speech than to that of the MT. These findings provide a neurological basis for separating the MT and the LT types, and support the widely-used MT/LT symptom grouping scheme. In addition, WWR play a similar role as the LT, and thus should be placed in the LT type.

When the Diagnosis Is Dual

When the Diagnosis Is Dual

Changes in the pattern of stuttering over development for children who recover or persist

Three schemes for assessing stuttering were compared. They differed with respect to whether they included whole-word repetitions as characteristics more typical of stuttering. Persistent and recovered groups of children were examined to see whether: (1) one of the schemes differentiated the groups better than others; (2) more and less typical of stuttering characteristics changed over ages in different ways for the groups; and (3) the changes over ages of more and less typical of stuttering classes depended on word type. Twenty-six children who stuttered were classified as persistent or recovered and recorded at three ages between 8–12+ years. Stuttering characteristics were obtained according to the three schemes. The results show: (1) All schemes distinguished persistent, from recovered, children. (2) The proportion of less typical of stuttering characteristics increased and more typical of stuttering characteristics decreased over age for recovered children for all three schemes (there was no change for persistent children). (3) The increased proportion of less typical of stuttering characteristics for the recovered children across age was more apparent for one-syllable content words than one-syllable function words.