No description available.

Background: Speech rate plays an essential role in overall speech intelligibility in fluent speakers and is an important variable affecting fluency in people who stutter (PWS). There are no normative speech rate data for South African English (SAE). In PWS, attempts to manipulate speech rate for improved fluency have mostly focused on articulation rate. Revisiting the role of pauses in speech rate and the manipulation of both frequency and duration of pauses was deemed necessary to investigate a potentially valuable alternative strategy to assist PWS in rate reduction and possible improved fluency. Aims: The aims of this study were to investigate the speech rate and pause use for first language (L1) and second language (L2) SAE fluent speakers and PWS in both monologue and reading tasks. In addition, this study investigated if the manipulation of pause use could increase fluency in L1 and L2 PWS given six sessions of pause instruction in SAE. Methods: 80 fluent speakers (40 L1 SAE and 40 L1 isiZulu) and 14 PWS (7 L1 SAE and 7 L1 isiZulu) were asked to engage in a 2-minute monologue and a reading task in order to calculate the mean speech rate, frequency of pauses and average pause duration for each group. Following baseline measures, the 14 PWS were randomly assigned to either immediate or delayed intervention consisting of 6 sessions addressing manipulation of pauses. A crossover treatment design allowed for repeated measures of speech rate and pause use across three data collection periods. Results: Results revealed L1 and L2 SAE fluent speakers differed significantly in speech rate and frequency of pauses in reading and in pause length in the monologue. L1 and L2 PWS differed in frequency of pauses in reading. L1 fluent speakers and PWS differed in speech rate and both measures of pause use in reading. L2 fluent speakers and PWS differed in frequency of pauses in the monologue. Results from the crossover intervention for the PWS revealed a significant decrease in percentage syllables stuttered (% SS) for the delayed treatment group and a clinically significant decrease in % SS for both groups, in conjunction with a reduced speech rate that could be linked to increased frequency and/or average duration of pauses. Conclusions: This study presented speech rate and pause use norms for both fluent speakers and PWS, L1 and L2 SAE speakers that may provide useful guidelines for speech language pathologists in South Africa. Additionally, the intervention results for PWS presented efficacy data for six sessions of pause manipulation with measurable findings for improved fluency.

O aspecto temporal da fala é um dos fatores que interferem na inteligibilidade de fala, assim, o presente estudo teve como objetivo estudar a percepção das vogais do português brasileiro (PB) em diferentes taxas de elocução por meio de análise acústica e metodologia psicofísica. Na análise acústica foi averiguado a duração das frases contendo as vogais alvo e, posteriormente, apenas as vogais. No experimento I - a ANOVA mostrou que existe diferença significativa apenas entre a duração das frases nas duas taxas de elocução normal e rápida, enquanto na análise da duração das vogais há diferença estatística entre as taxas de elocução e também entre as vogais. No experimento II- o resultado da ANOVA mostrou que as frases são estatisticamente diferentes em relação a taxa de elocução, normal e mais rápida, mas não difere quanto as diferentes vogais alvo das frases, entretanto, o resultado das vogais na taxa de elocução mais rápida mostrou que há diferença entre as mesmas. A análise dos três conjuntos de vogais dos dois experimentos anteriores, mostrou que há diferença estatisticamente significativa nas taxas de elocução e também nas 7 vogais do PB. O julgamento psicofísico da fala comprimida foi realizado por meio de 28 frases aleatorizadas quanto as 7 vogais do PB, duas taxas de locução (normal e rápida) e duas apresentações para fidedignidade do teste. A amostra foi composta de 32 sujeitos, distribuídos em dois experimentos. No experimento I (n=16) os sujeitos não perceberam qualquer dificuldade na inteligibilidade de fala. No segundo experimento (n=16) a taxa de elocução das frases rápidas foi modificada em torno de 30% da taxa normal. O resultado mostrou que os sujeitos perceberam diminuição na inteligibilidade de fala devido a taxa de elocução, entretanto não encontraram qualquer dificuldades devido as vogais. Os resultados sugerem a taxa de elocução interfere na inteligibilidade de fala. / The temporal aspect of speech is one of the factors that affect the intelligibility of speech. Thus, the aim of the present study was to investigate the perception of Brazilian Portuguese (BP) vowels in different speech rates through acoustic analysis and psychophysical tests. In the acoustic analysis, we investigated the sentences duration containing target vowels and, subsequently, only the vowels. In Experiment I, there was a significant difference between the duration of sentences in both speech rates (normal and fast), and in the duration of the vowels there was a significant difference between the speech rates and between the vowels. In Experiments II, the results showed differences between the sentences concerning the speed rate, normal and 'faster', but not between the target vowels in the sentences. However, in the 'faster' speech rate showed difference between the vowels. The analyses of the 3 sets of vowels in both experiments revealed a significant difference in speed rates as well as in the 7 vowels of the BP. The psychophysical judgment of compressed speech was done using 28 randomized sentences with 7 BP vowels, 2 speed rates (normal and fast) and 2 repetitions (test and retest) to evaluate the test reliability. The sample was composed by 32 subjects distributed in 2 experiments. In Experiment 1 (n=16) the subjects did not have difficulty in the speech intelligibility In Experiment II (n=16) the speed rate of 'faster' sentences was increased in 30% in relation to the normal rate. The results showed that the subjects noticed that there was a decrease in the speech intelligibility due to the speed; however, they did not find difficulties due to the vowels. Taken together, our results suggest that the speech rate interferes in the speech intelligibility.

While a number of studies have sought to investigate ESL speakers' fluency gains over the course of one 15-week semester, few if any studies have investigated these changes over a longer developmental period. A critical factor in researching longitudinal change is that students do not often remain in an intensive English program (IEP) for more than two semesters before moving to a new school, applying to an American university, or returning to their home country. Longitudinal research, therefore, is necessary as program administrators, teachers, and learners all seek to understand points where change in oral fluency actually occurs. For this study data were collected from students in a large intensive English program over a 45-week period. For 39 ESL learners audio files from speaking tasks that were part of placement and end-of-semester level achievement tests were collected and analyzed. Specific oral fluency features such as speech rate, articulation rate, and pause frequency were investigated. This thesis will share the results of the analysis while also discussing the implications of the data for program administrators, teachers, and learners. Particular focus will be given to helping stakeholders understand specific changes that occurred in learners' fluency over the time period of three semesters.

L’objectif de cette thèse est de modéliser l’interlangue prosodique des apprenants francophones de l’anglais afin de pouvoir élaborer des critères utilisables pour une évaluation semi-automatique de leur niveau prosodique. Le domaine évaluatif requiert la plus grande rigueur dans la mise en place de ses critères pour aboutir à la validité, la fiabilité, la faisabilité et l’équité maximales, alors que la prosodie anglaise de la langue cible se caractérise par son extrême variabilité. Aussi, peu d’études se sont engagées dans l’évaluation de la prosodie, qui représente une réelle gageure. Pour relever ce défi, une stratégie particulière a été mise en place pour élaborer une méthodologie permettant d’atteindre l’objectif fixé, en lecture.L’approche choisie repose sur la symbiose permanente qu’entretient la prosodie avec le monde dans lequel évolue le locuteur. Cette méthodologie, ou « profilage », est destinée à sélectionner par inclusion ou exclusion les éléments analysés tant au niveau perceptif qu’acoustique. Le profilage des réalisations sur l’axe syntagmatique permet de sélectionner les locuteurs natifs servant de modèles, et celui basé sur le phénomène d’emphase rend possible un ciblage de leurs réalisations les plus pertinentes à modéliser sur l’axe paradigmatique. Conformément à cette méthodologie d’investigation nouvelle et aux résultats perceptifs et acoustiques concordants pour la langue cible, les réalisations des apprenants francophones du corpus Longdale-Charliphonia sont analysés acoustiquement. Le classement automatique à partir des variables prosodiques (acoustiques et perceptives) est confronté à celui d’experts évaluant par perception classique.Les travaux de cette thèse aboutissent essentiellement à : Une modélisation de la prosodie anglaise non native par grilles évaluatives critériées s’appuyant sur critères distinctifs natifs et non natifs issus de variables temporelles (vitesse d’élocution avec ou sans pauses), de registre et de mélodie, ainsi que de rythme, À partir de ces variables, une évaluation semi-automatisée de 15 apprenants représentatifs du corpus par classement et notation, une correspondance des résultats de l’évaluation traditionnelle avec celle semi-automatique évoluant entre 56,83% et 59,74% dans une catégorisation des apprenants en 3 niveaux de maîtrise, en fonction du profilage d’experts évaluateurs. / The aim of our study is to modelise the prosodic interlanguage of Francophone learners of English in order to provide useful criteria for a semi-automatic assessment of their prosodic level in English. Learner assessment is a field that requires to be very rigorous and fair when setting up criteria that ensure validity, reliability, feasibility and equality, whereas English prosody is highly variable. Hence, few studies have carried out research in assessing prosody because it represents a real challenge. To address this issue, a specific strategy has been devised to elaborate a methodology that would ensure assessing a reading task successfully.The approach relies upon the constant symbiosis between prosody and a speaker’s subjective response to their environment. Our methodology, also known as « profiling », first aims at selecting relevant native perceived and acoustic prosodic features that will optimize assessment criteria by using their degree of emphasis and creating speakers’ prosodic profiles. Then, using the Longdale-Charliphonia corpus, the learner's productions are analysed acoustically. The automatic classification of the learners based on acoustic or perception prosodic variables is then submitted to expert aural assessment which assesses the learner evaluation criteria.This study achieves: A modelisation of non-native English prosody based on assessment grids that rely upon features of both native and non-native speakers of English, namely, speech rate – with or without the inclusion of pauses, register, melody and rhythm,A semi-automatic evaluation of 15 representative learners based on the above modelisation – ranking and marking,A comparison of the semi-automatic results with those of experts' auditory assessment; correspondence between the two varies from 56.83% to 59.74% when categorising the learners into three prosodic proficiency groups.

<p>Recordings of speech play a central role in the diverse subdisciplines of linguistics. The reliance on speech recordings is especially profound in sociolinguistics, where scholars have developed a range of techniques for eliciting and analyzing natural talk. Despite the focus on naturalistic speech data, sociolinguists have rarely focused explicitly on the management (e.g. organization, storage, accessibility, and preservation) of their data, and this lack of focus has had consequences for the advancement of the field. At the same time, the interviews that sociolinguists labor so hard to obtain are often barely mined for their full potential to further our understanding of language. That is, sociolinguists often focus on a handful of phonological and/or morphosyntactic variables to the exclusion of so many other features of speech. The present work both addresses the management of sociolinguistic data and, through an innovative approach to speech data management and analysis, extends the sociolinguistic lens to include the lesser-examined realm of variation in sequential temporal patterns of talk.</p><p>The first part of this dissertation describes the Sociolinguistic Archive and Analysis Project (SLAAP), a web-based digitization and preservation initiative at North Carolina State University. SLAAP, which I principally have designed and developed, is more than an archive; it has actively sought to explicate approaches to spoken language data management and to enrich spoken language data through the development of analytic tools designed specifically for sociolinguistic analysis. This dissertation begins by situating SLAAP within the history of data management practices in the field of sociolinguistics. It then provides an overview of many of SLAAP's features, discussing in particular the transcript model that enables most of its analytic and presentational capabilities.</p><p>The second part of this dissertation takes advantage of SLAAP's data model and the extensive language data accumulated within its archive to examine variation in speech rate and silent pause duration by North American English speakers of four ethnicities in North Carolina, Ohio, Texas, Washington, DC, and Newfoundland. This work brings a wide range of previous research from different areas of sociolinguistics, psycholinguistics, and corpus linguistics to bear on an array of quantitative analyses, demonstrating that speech rate and pause exhibit meaningful variation at the social level at the same time as they are also constrained by cognitive and articulatory processes. </p><p>Specifically, pause and speech rate are shown to vary by region, ethnicity, and gender - albeit not in mono-directional ways - although other factors arise as significant, including, for speech rate, a strong effect of utterance length as well as a number of interactional or discourse-related factors, such as the gender of the interviewer and the number of participants in the speech event. A number of the examinations undertaken relate sociolinguistic conceptions of style to language production and cognitive processes, including a quantitative analysis of sequential temporal patterns as paralinguistic cues to attention to speech, performativity, and the realization of phonological and morphosyntactic variables. Through this analysis, sociolinguistic data and findings are brought to bear on a tradition of psycholinguistic investigations with the hope to benefit both, often disparate, areas of research.</p> / Dissertation

Tal- och röstförändringar är vanligt förekommande hos personer med Parkinsons sjukdom. Dessa påverkar ofta talarens förståelighet men kan också ha en negativ inverkan på talets naturlighet. Forskning angående vilka faktorer som påverkar talets naturlighet är i dagsläget begränsad, varför föreliggande studie har genomförts. Syftet med studien var att undersöka huruvida den uppfattade talnaturligheten kunde härledas till några specifika tal- och röstparametrar. I föreliggande studie konstruerades ett testbatteri för att elicitera talmaterial från åtta personer med Parkinsons sjukdom. Forskningspersonernas röster spelades in och inspelningarna graderades sedan av 27 naiva lyssnare gällande förståelighet och talnaturlighet. Korrelationstester genomfördes slutligen för att hitta eventuella samband mellan lyssnarnas bedömning och olika akustiska parametrar. Resultatet visade att tal- och artikulationshastighet var den faktor med störst inverkan på lyssnargruppens bedömning av talnaturlighet, där de med långsammast hastighet bedömdes ha mest onaturligt tal. Vidare fanns starka indikationer på att grad av förståelighet korrelerade med bedömningen av talnaturlighet. I föreliggande studie tycktes inga övriga akustiska parametrar ha en statistiskt signifikant korrelation med lyssnargruppens bedömning av talnaturlighet. / Speech and voice changes are common in Parkinson’s disease. These changes can affect the speaker’s intelligibility but can also have a negative impact on the perceived naturalness of speech. The research available regarding the different factors that affect speech naturalness is scarce, which was the motivation behind this study. The aim of the present study was to investigate whether the level of perceived speech naturalness could derive from any specific aspects of speech. This was accomplished by recording speech samples from eight people with Parkinson’s disease using a test battery with various speech tasks. These samples were presented to a group of 27 naive listeners whose task was to judge the level of intelligibility as well as the level of speech naturalness. Correlations were then made between their assessments and various acoustic measurements. The main finding of the present study was that speech and articulation rate seemed to have the greatest impact on the perceived level of naturalness, where the people who had the slowest rate were judged to be the least natural sounding. Furthermore there were strong indications that the level of intelligibility correlated with the level of speech naturalness. In this study there were no other acoustic correlates found with statistical significance.

O aspecto temporal da fala é um dos fatores que interferem na inteligibilidade de fala, assim, o presente estudo teve como objetivo estudar a percepção das vogais do português brasileiro (PB) em diferentes taxas de elocução por meio de análise acústica e metodologia psicofísica. Na análise acústica foi averiguado a duração das frases contendo as vogais alvo e, posteriormente, apenas as vogais. No experimento I - a ANOVA mostrou que existe diferença significativa apenas entre a duração das frases nas duas taxas de elocução normal e rápida, enquanto na análise da duração das vogais há diferença estatística entre as taxas de elocução e também entre as vogais. No experimento II- o resultado da ANOVA mostrou que as frases são estatisticamente diferentes em relação a taxa de elocução, normal e mais rápida, mas não difere quanto as diferentes vogais alvo das frases, entretanto, o resultado das vogais na taxa de elocução mais rápida mostrou que há diferença entre as mesmas. A análise dos três conjuntos de vogais dos dois experimentos anteriores, mostrou que há diferença estatisticamente significativa nas taxas de elocução e também nas 7 vogais do PB. O julgamento psicofísico da fala comprimida foi realizado por meio de 28 frases aleatorizadas quanto as 7 vogais do PB, duas taxas de locução (normal e rápida) e duas apresentações para fidedignidade do teste. A amostra foi composta de 32 sujeitos, distribuídos em dois experimentos. No experimento I (n=16) os sujeitos não perceberam qualquer dificuldade na inteligibilidade de fala. No segundo experimento (n=16) a taxa de elocução das frases rápidas foi modificada em torno de 30% da taxa normal. O resultado mostrou que os sujeitos perceberam diminuição na inteligibilidade de fala devido a taxa de elocução, entretanto não encontraram qualquer dificuldades devido as vogais. Os resultados sugerem a taxa de elocução interfere na inteligibilidade de fala. / The temporal aspect of speech is one of the factors that affect the intelligibility of speech. Thus, the aim of the present study was to investigate the perception of Brazilian Portuguese (BP) vowels in different speech rates through acoustic analysis and psychophysical tests. In the acoustic analysis, we investigated the sentences duration containing target vowels and, subsequently, only the vowels. In Experiment I, there was a significant difference between the duration of sentences in both speech rates (normal and fast), and in the duration of the vowels there was a significant difference between the speech rates and between the vowels. In Experiments II, the results showed differences between the sentences concerning the speed rate, normal and 'faster', but not between the target vowels in the sentences. However, in the 'faster' speech rate showed difference between the vowels. The analyses of the 3 sets of vowels in both experiments revealed a significant difference in speed rates as well as in the 7 vowels of the BP. The psychophysical judgment of compressed speech was done using 28 randomized sentences with 7 BP vowels, 2 speed rates (normal and fast) and 2 repetitions (test and retest) to evaluate the test reliability. The sample was composed by 32 subjects distributed in 2 experiments. In Experiment 1 (n=16) the subjects did not have difficulty in the speech intelligibility In Experiment II (n=16) the speed rate of 'faster' sentences was increased in 30% in relation to the normal rate. The results showed that the subjects noticed that there was a decrease in the speech intelligibility due to the speed; however, they did not find difficulties due to the vowels. Taken together, our results suggest that the speech rate interferes in the speech intelligibility.

The purpose of this study was to examine the following prosodic elements: speech rate, turn-taking latency, number of interruptions, and utterance duration across two groups' neurotypical and autistic young adults. Furthermore, the end goal of this study is to help provide a baseline and clinical application of prosodic differences between autistic and neurotypical adults. Speech samples were collected from 11 neurotypical and 11 autistic young adults from the ages of 18-26. Speech samples were recorded responses from a 10-minute interview between two research assistants and the autistic or neurotypical individual. Using Praat software, speech samples were analyzed and used to calculate speech rate, utterance duration, turn-taking latency, and the number of interruptions for each subject. Across the four prosodic elements, there were significant differences between the autistic and neurotypical groups. The neurotypical group exhibited significantly higher speech and interruption rates when compared to the autistic group. Whereas, the autistic group displayed longer turn-taking latency periods and longer utterance durations. Across all conditions, there were no significant difference between biological sex or effect of familiarity within the autistic and neurotypical groups. Results of this study provide clinicians and researchers a baseline of prosodic differences found between autistic and neurotypical individuals. Future research is needed to better understand how these findings might improve the assessment and treatment of autistic individuals.

Orientador: Prof. Dr. Francisco J. Fraga / Dissertação (mestrado) - Universidade Federal do ABC, Programa de Pós-Graduação em Engenharia da Informação, 2014. / Most speaker verifcation systems are based on Gaussian mixture models and more recently on the so-called i-vector. These two methods are affected in mismatched testtrain conditions, which might be caused by vocal-efort variability, different speakingstyles or channel efects. In this work, we compared the impact of speech rate variation and room reverberation on both methods. We found that performance degradation due to variation on speech rate can be mitigated by adding fast speech samples into the training set, which decreased equal error rates for Gaussian mixture models and i-vector, respectively. Regarding reverberation, we investigated the achievements of both methods when three diferent reverberation compensation techniques are applied in order to overcome performance degradation. The results showed that having reverberant background models separated by diferent levels of reverberation can bene t both methods, with the i-vector providing the best performance in that scenario. Finally, the performance of two auditory-inspired features, mel-frequency cepstral coe ficients and the so-called modulation spectrum features, are compared in presence of room reverberation. For the speaker verifcation system considered in this work, modulation spectrum features are equally afected by reverberation time and have their performance degraded as the level of reverberation increases.