Thesis (Ph. D.)--University of California, San Diego, 1998. / Vita. Includes bibliographical references (leaves 238-249).

Thesis (M.A.)--Brandeis University, 2009. / Title from PDF title page (viewed on June 29, 2009). Includes bibliographical references.

Thesis (Ph. D.)--UCLA, 2009. / Vita. Description based on print version record. Includes bibliographical references (leaves 112-123).

Thesis (Ph.D.)—Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1981. / Includes bibliographical references (leaves 185-192). / This electronic version was scanned from a copy of the thesis on file at the Speech Communication Group. The certified thesis is available in the Institute Archives and Special Collections. / Whitaker Health Sciences Fund.

El objetivo del siguiente artículo es presentar un estudio sobre las vocales susurradas, y su gradual desaparición, en la lengua shuar. El estudio parte de la premisa de que las vocales susurradas desaparecen de la lengua shuar por la acomodación al sistema vocálico del español. Este estudio se fundamenta en la investigación realizada en 21 comunidades a lo largo del territorio shuar. Se examinan tres factores extralingüísticos: edad, localidad y grado del bilingüismo. La conclusión confirma que el contacto con el español juega un papel importante en la desaparición de las vocales susurradas porque la intensidad del contacto es el factor que más influye en su desuso gradual. / The objective of the following paper is to present a study about the voiceless vowels in Shuar and their decreasing vitality. The hypothesis of the study presumes that the voiceless vowels disappear due to the influence of Spanish vocalic system. The database for the research was collected in 21 Shuar communities. The study takes into account three extralinguistic factors: age, locality and grade of bilingualism. The conclusion confirms that the contact with Spanish plays an important role in the disappearance of the voiceless vowels as the intensity of contact with Spanish is the most influential factor in their gradual disuse.

Orientador: Eleonora Cavalcante Albano / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Estudos da Linguagem / Made available in DSpace on 2018-08-16T00:29:50Z (GMT). No. of bitstreams: 1 Roces_Laudino_D.pdf: 1998697 bytes, checksum: 7d43c695710ab723ef54f6ca095719c0 (MD5) Previous issue date: 2010 / Resumo: Esta pesquisa teve por finalidade estudar as relações formânticas vogal -a - vogal em sequências V'CV do Português Brasileiro, descritas na literatura como coarticulação ou coprodução V-V. O corpus do trabalho foi composto de pares V-V, com as vogais /i,e,a,o,u/ na posição pré-tônica e as vogais /i,e,E, a, ç, o,u/ na posição tônica, com dados de 20 falantes do Português Brasileiro da região sudeste. As análises dos dados foram divididas e quantitativas e qualitativas. Nas quantitativas, utilizamos os valores acústicos da totalidade do corpus e nas qualitativas efetuamos testes individuais nos sujeitos do corpus e, a partir daí, tabulamos os resultados, passando a tratá-los qualitativamente. Estas duas categorias de análise possibilitaram a reunião de uma série de elementos sobre a descrição das relações coarticulatórias V-V em sequências VCV, algumas delas inéditas na literatura. Observamos que as relações formântica V-V apresentam não só a possibilidade deassimilar a configuração formântica da vogal adjacente, mas, também a possibilidade de dissimilar, ou seja, da influência da vogal adjacente fazer com que os formantes aumentem ou diminuam seu valor no sentido contrário ao que se esperaria na assimilação. Outro elemento importante do estudo foi a introdução de uma variável visando o controle da fala hiperarticulada, ou fala clara, A diferença entre as sequências com e sem foco foram significativas nas análises quantitativas. A duração para as sequências com foco foram sempre maiores. Medimos também a diferença dos entre F1 e F2, chamando-as Delta V-V. Nas associações entre os processos de assimilação/dissimilação e Delta V-V, os resultados dos testes apontam para uma relação significativa entre a presença de efeitos de assimilação/dissimilação e a ocorrência de comportamento previsível em Delta V-V em F2. Os resultados das análises individuais utilizando o modelo GLM mostram que os sujeitos, em geral, apresentaram resultados tanto para os processos de assimilação quanto para os processos de dissimilação. O estudo mostrou que não se pode esperar um comportamento gradiente completamente previsível dos processos fônicos V-V, e, sim, tendências e direcionamentos, sob a influência de uma série de variáveis / Abstract: This work, studied the relations V-V in in Brazilian Portuguese VCV sequences. The literature about coarticulation classifies the coarticulation V-V in perseveratory or carryover assimilations, however, we found perseveratory dissimilation effects. The qualitative tests shows that the subjects, in general, presented resulted so much for the assimilation processes as for the dissimilation processes, although, we cannot say that all the subjects make use of the dissimilation process in yours strategies. The percentage of assimilation occurrence was of 21% in F1, 34% in F2 and 21% in F3 and for the dissimilation it was of 15% in F1, 4% in F2 and 9% in F3. There is, therefore, predominance of dissimilation. The quantitative tests pointed the relationship of assimilation/dissimilation processes with the stressed position (V2). In V1 (prestressed) all vowels received significant influences of the adjacent vowel (V2). In the stressed position, the vowels /i/ and /u/ didn't show significant results. The quantitative tests showed that the assimilation has a preference for the prestressed position (V1), while the dissimilation only occurs in the stressed position, despite the great susceptibility of the vowels in prestresse position to coarticulatory influences pointed in literature. The qualitative studies have confirmed that preference, showing a higher incidence of significant associations between assimilation and preaccentual position and the occurrence of some significant associations between the dissimilation process and the accentual position. The result shows that the process of dissimilation occurs preferentially in F1. This preference can be explained by another important finding of this study: indications of a compensation process in the stress position. Some of the values in Hz for F1 of mid vowels are lower in data with focus rather than the data without focus. Data without focus have lower durations when compared with data with focus. According to the literature more time would necessarily have to be accompanied by a rise in the value of F1. The relationship between the two types of compensation - found points to a complex strategy to restore the quality of the vowel despite the opening of the jaw associated processes is V-V. Thus where the vowel should be kept intelligibility, the longest or coproduction of segments could cause the obliteration of some distinctions. The retraction of the root of the tongue offset these effects in the mid vowels, which are relatively close together in acoustic-auditory terms. Finally, the variability of the behavior of the processes indicated in this study points to explanations based on dynamic models of speech production, in which fluctuations and compensation are the norm rather than exception. The study shows that one can't expect a completely predictable behavior gradient vowel-to-vowel processes, but trends and directions under the influence of several variables whose behavior we could not control in the experiments / Doutorado / Linguistica / Doutor em Linguística

M.A. (African Languages) / The phonological system of Zulu does not tolerate consecutive vowels in one word while, with specific conditions, only certain consonant combinations and consonant-vowel combinations are permissible. Morphonological adjustments such as glide insertion, vowel coalescence and semi vocalization are therefore utilized to avoid the accumulation of vowels. Where consonants are brought together by morpheme combinations, morphonological adjustments such as nasalization occur to avoid the impermissible consonant groups. Adjustments such as palatalization are the result of the impermissible combinations of certain consonants and (one or more) vowels and/or semivowels. This study aims at analysing some known morphonological phenomena in Zulu from a generative-phonological point of view. It involves mainly the construction and formulation of a number of morphonological rules for Zulu within the framework of the transformational generative (T.G.) phonology. The study is purely introductory and does not claim that all relevant material has been fully analysed and exhausted. The point of departure is that of Chomsky and Halle in their Sound Patterns in English (SPE). Chapter one defines the concept of phonological processes, states the conditions for acceptable phonological rules and supplies definitions of features employed to describe the segments of Zulu. Chapter two identifies the underlying segments of Zulu, i. e. vowels, consonants, "impure" consonants (so-called liquids) and semi-vowels. Characteristics of the underlying segments are presented in a table. Through the formulation of redundancy rules those features of segments which systematically correspond with other features are eliminated and lexical versions of the underlying segments are determined. In chapter three the processes of glide vowel coalescence which relate to insertion, semi vocalization an impermissible series of vowels are discussed and summarized in rules.

Cette recherche veut unir deux domaines qui pour longtemps on été séparés : la dialectologie et la phonétique expérimentale et, bénéficiant de ce rapprochement, elle voudrait apporter sa contribution à une connaissance plus approfondie des dialectes du Salentin (Pouilles, Italie du sud). Nous avons analysé acoustiquement et articulatoirement certains aspects encore peu connus de certaines variétés du Salentin central (Lecce, Monteroni di Lecce, Nardò, Squinzano et Torchiarolo): le vocalisme atone et tonique et ses caractéristiques (les diphtongues métaphoniques palatale et labio-vélaire /wƐ/ et /jƐ/), ainsi que les consonnes rétroflexes. Plus spécifiquement, nous avons étudié la réalisation acoustique des voyelles toniques, en particulier, celles des voyelles moyennes antérieure et postérieure /Ɛ/ et /ɔ/ quand elles sont suivies des contextes finaux atones -i, -u et -e/-a/ ; l’action métaphonique des voyelles hautes atones finales -i et -u sur les voyelles moyennes toniques antérieure et postérieure et les processus de changement phonétique à l’origine de rétroflexes au niveau des liquides latérales (géminées en position intervocalique) et vibrantes (dans des groupes homorganiques tautosyllabiques), dans cette zone romane.Les résultats de notre analyse acoustique montrent que le vocalisme tonique de ces cinq points d’enquête est asymétrique, c’est-à-dire que la voyelle moyenne antérieure /Ɛ/ est plus fermée que la voyelle moyenne postérieure /ɔ/; le vocalisme atone ne déclenche d’action métaphonique que pour Monteroni di Lecce où la voyelle /Ɛ/ devient /e/ quand elle est suivie de la voyelle haute finale atone –i. La métaphonie trouvée par Grimaldi (2003) dans le salentin méridional est très répandue au sud extrême du Salento et au fur et à mesure que l’on va vers le nord ce processus s’estompe progressivement : nous ne nous attendions donc pas à trouver un effet de ce type dans cette zone. En ce qui concerne les rétroflexes, d’après nos analyses acoustiques, articulatoires et la littérature précédente, la transcription I.P.A. qui nous semble plus appropriée pour ces segments est [ḍ᷇z] pour le reflet de la latérale latine géminée -LL- qui est un segment cacuminal, géminé (durée de consonne plutôt long), semi-affriqué (burst plus long que pour une simple occlusive et présence de bruit de friction) et alvéolaire/post-alvéolaire (valeur du locus, limite inférieure du bruit, valeur du CoG et électrodes activées dans l’étude éléctropalatographique); les groupes consonantiques [ṭṣ__] et [ṭ :ṣ__]sont des segments cacuminaux, simple et long respectivement (durée totale des groupes consonantiques), affriqués (bruit de friction plutôt long) et alvéolaires/post-alvéolaires (sur la base de la valeur du locus, la limite inférieure du bruit et la valeur du CoG et les électrodes activées dans l’étude éléctropalatographique). / This study makes an attempt to unify two fields which have been separated for a long time : dialectology and experimental phonetics. Benefiting from this approach, our research aims at contributing to provide a deeper knowledge of Sallentine dialectes (Apulia, southern Italy).We analysed acoustically and articulatory some unknonwn aspects of some varieties of central Sallentine (Lecce, Monteroni di Lecce, Nardò, Squinzano et Torchiarolo): unstressed and stressed vocalism and its characteristics (metaphonical palatal and labio-velar diphtongues / wƐ/and /jƐ/ ), as well as retroflex consonants. Specifically, we studied the acoustic realisations of stressed anterior and posterior vowels /Ɛ/ / and /ɔ/ when they are followed by final unstressed vowels -i, -u and -e/-a/; the metaphonic action of unstressed vowels -i and -u on stressed anterior and posterior vowels as well as the phonetic processes concerning lateral liquids (geminate in intervocalic position) and trills (in homorgamical tautosyllabic groups) in this Romance area. Acoustical results show that stressed vocalism at all research points is asymmetric, the anterior vowel /Ɛ/ being closer than the posterior one /ɔ/; unstressed vocalism causes metaphonic action only in Monteroni di Lecce where /Ɛ/ becomes /e/ when it is followed by a final unstressed vowel –i. As metaphony found by Grimaldi (2003) in the southern Sallentine is present in the extreme South of Salento but seems to disappear progressively towards the North, we didn’t except to find this type of effect in this zone. About retroflexes, following our acoustical, articulatory analysis and previous literature, appropriate I.P.A. transcription for these segments is ḍ᷇z] for Latin lateral geminate output -LL- which is a cacuminal, geminate segment (consonant total duration), semi-affricate (longer burst than in plosive and presence of friction noise) and alveolar/post-alveolar (based on locus value, inferior noise limit, CoG value and activated electrodes in the electropalatographic study); consonantal clusters [ṭṣ__] and [ṭ :ṣ__] are cacuminal, simple and long segments respectively (consonant total duration), affricative (long burst) and alveolar/post-alveolar (locus value, inferior noise limit, CoG value and activated electrodes in the electropalatographic study).

This thesis focuses on the production and perception of ten English vowels (/i, ɪ, e, ɛ, æ, ʌ, ɑ, o, ʊ, u/) by native speakers of Brazilian Portuguese (BP) living in Victoria, British Columbia, Canada. The participants consisted of 14 native speakers of BP (divided into intermediate and advanced second language (L2) English proficiency groups), plus six native speakers of Canadian English (CE) as control participants. Four experiments were carried out: two pertaining to production and two pertaining to perception. The production tasks consisted of CE and BP wordlist reading tasks in order to measure the duration and first two formants of the participants’ vowels, while the perception tasks consisted of an identification task and an oddity-discrimination task. With regards to production, this thesis investigates how the participants’ productions of the L2 vowels differ between the experimental and control groups with respect to their formant frequencies and the Euclidean distances (EDs) between various English vowel pairs. Similarly, the participants’ perceptual abilities, as measured by their performance on the perception tests, are examined. Finally, the connection between perception and production is investigated. The findings indicate (a) a positive effect of proficiency, as the advanced participants showed a greater ability to both produce and perceive the L2 vowels, but that (b) participants of both proficiency levels have difficulty in contrasting certain English vowel pairs in a native-like fashion. Furthermore, the findings provide evidence of a connection between perception and production, and show that perception outperforms production, as predicted by the Speech Learning Model (Flege, 1995, 2005). Finally, the findings indicate a possible positive effect of environment (i.e., an English-speaking country) when compared to previous studies (Bion et al., 2006, Rauber, 2006). Pedagogical implications of these findings are also discussed. / Graduate

This study examined changes in speech production when distracting background audio is present. Forty typically speaking adults completed a repetitive sentence reading task in the presence of 5 different audio conditions (pink noise, movie dialogue, heated debate, classical music, and contemporary music) and a silent condition. Acoustic parameters measured during the study included vowel space area (VSA), vowel articulation index (VAI), formant transition extent, formant transition rate, and diphthong duration for /ɑɪ/ and /ɑʊ/. It was hypothesized that there would be significant increases in vowel space area and vowel articulation index as well as an increase in formant transition measures in the presence of background noise. There were statistically significant decreases in vowel space are and vowel articulation index in the presence of all noise conditions compared to the silent baseline condition. Results also demonstrated a significant decrease in F2 transition extent for both /ɑɪ/, and /ɑʊ/ diphthongs in all noise conditions except the pink noise condition when compared to the silent condition. These findings were contrary to what was originally hypothesized. It is possible that VAI and VSA decreased in the presence of background noise due to an increase in speaking rate. Formant transition measurements were consistent with the VAI and VSA results. More research is needed to accurately determine the acoustic changes a speaker makes in response to distracting background audio.