Orientador: Eleonora Cavalcante Albano / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Estudos da Linguagem / Made available in DSpace on 2018-08-21T00:01:21Z (GMT). No. of bitstreams: 1 Meneses_Francisco_M.pdf: 3600241 bytes, checksum: 3773bdafbef97346d85bb068f8bd4a65 (MD5) Previous issue date: 2012 / Resumo: Este trabalho examina o processo de produção de vogais desvozeadas no português brasileiro (doravante PB). O objetivo é, a partir de uma visão de cunho dinâmico, mostrar que há, no sinal das chamadas sílabas desvozeadas, rastros da vogal que impossibilitam uma hipótese de síncope vocálica. Além disso, busca-se uma síntese teórica das pistas acústicas encontradas a fim de relacioná-las à articulação das vogais desvozeadas. Para tanto, uma análise acústico-articulatória foi realizada a partir da gravação da leitura de frases-veículos. As leituras das frases foram realizadas por seis sujeitos do sexo feminino, naturais de Vitória da Conquista - BA, sem queixas de fala ou audição. As gravações foram realizadas em uma cabine acústica, por meio de gravador digital. A análise instrumental foi realizada por meio do software PRAAT. Foram obtidas as medidas de duração da sílaba e do ruído fricativo, as medidas do centroide do ruído e os valores da Razão de centralização formântica (doravante FCR) e a Área de espaço vocálico (doravante VSA) das vogais em contexto de desvozeamento. Os resultados encontrados mostram que há um gradiente de desvozeamento manifestado de três maneiras: as medidas de duração, assim como as medidas do primeiro momento espectral, mostraram que pistas remanescentes do gesto vocálico permanecem no ruído das fricativas; os dados de FCR e VSA mostram que as vogais sofrem uma grande redução da magnitude em contexto de desvozeamento. Os resultados apontam para um efeito de sobreposição de gestos, em detrimento de uma síncope vocálica. Essa hipótese não caberia em descrições fonológicas tradicionais, as quais, em geral, lidam com operações simbólicas. O fenômeno em estudo pode, então, ser iluminado pela Fonologia Gestual (BROWMAN E GOLDSTEIN, 1992; BALL E KENT, 1997; ALBANO, 2001), a qual, em relação à representação tradicional, é capaz de expressar realizações gradientes, pois incorpora com sucesso os fatores tempo e magnitude, diretamente relacionados à ideia de movimento dos articuladores / Abstract: This paper examines the production of devoiced vowels in Brazilian Portuguese. The goal is to depart from a dynamic view to show that there are vowel traces in the signal of devoiced syllables, a fact which weakens the hypothesis of vowel syncope. In addition, we seek a theoretical synthesis of the acoustic cues found in order to relate them to devoiced vowel articulation. To this end, an acoustic-articulatory analysis was performed from the recording of read carrier sentences. The sentences were read by six female subjects, from Vitória da Conquista - BA, without speech or hearing problems. The recordings were performed in an acoustic booth, using a digital recorder. The instrumental analysis was performed using the PRAAT software. The following measurements were made: the duration of the syllable and its fricative noise, the centroid of the noise and the values of formant centralization ratio (FCR) and vowel space area (VSA) of vowels in devoicing context. The results show that there is gradient devoicing manifested in three ways: both the duration measurements and the first spectral moment measures showed that the vowel gesture cues remain in fricative noise signal; FCR and VSA data show that vowels undergo great magnitude reduction in the devoicing context. The results indicate an effect of overlapping gestures, rather than vowel syncope. This hypothesis would not fit into traditional phonological descriptions, which usually deal with symbolic operations. The phenomenon under study can thus then be illuminated by Gestural Phonology (BROWMAN AND GOLDSTEIN, 1992; BALL AND KENT, 1997; ALBANO, 2001), which, as opposed to traditional representation, is able to express gradients, as it incorporates successfully the factors of timing and magnitude, directly related to the idea of articulator movement / Mestrado / Linguistica / Mestre em Linguística

Mental imagery is a common theme in research that clarifies how musical thought relates to musical performance. Unfortunately, minimal information exists regarding mental imagery and singers. The purpose of this study was to probe the role, if any, mental imagery plays in the conceptualization of pitch and vowel. By interviewing singers at differing levels of expertise, basic information was obtained about the mental processes used by singers. Through evaluations of the singers' mental processes, it was concluded that 95% of the singers in the study employed mental imagery. All singers described using kinesthetic imagery, while the majority implemented sensory and auditory imagery. Viso-spatial imagery was implemented among the more experienced singers. The majority of singers also reported: imaging pitch and vowel interactively; imaging from an internal perspective; and utilizing mental rehearsal. Less than half of the singers described using methods other than mental imagery to conceptualize pitch and vowel.

Pronunciation problems of Czech French-learning students and the effectiveness of learning strategies applied by the students of Faculty of Education, Charles University and Faculty of Philosophy and Arts, University of West Bohemia AUTHOR: PhDr. Kateřina Juřičková DEPARTMENT: Department of French Language and Literature, Charles University TUTOR: doc. PhDr. Marie Fenclová, CSc. This PhD thesis - a comparative study - deals on the segmental level with vowels which Czech French-learning speakers find difficult to pronounce and on the suprasegmental level with prosodic factors. On the former level, closed vowels both labialized and sharp, nasal vowels, schwa and semi-vowels are analysed. The first, theoretical part of the paper contains the phonetic and phonological characteristics of the Czech language and the French language from the viewpoint of vocalic systems of these languages and from the viewpoint of melody-prosody interaction. The second, experimental part, contains descriptions of the pre- research single phase observation of problematic phenomena obtained from three groups of students of KFJL (Department of French Language and Literature; Faculty of Education - Charles University), and findings from the research proper from two-phase observations of two groups of students of KFJL and...

This thesis explores the challenges faced by German native speakers in perceiving and producing French nasal vowels. French has three to four nasal vowels, while German lacks them entirely. The study analyzes German speakers of different proficiency levels using a perception task and production measurements. The results indicate that German speakers struggle in perceiving nasal contrasts, but can perceive nasality in itself. Acoustic analysis reveals deviations in nasalance and formant values compared to native French speakers. However, L2 speakers successfully master other markers of nasality. The findings highlight the interplay between perception and production, impacting the acquisition of nasal vowels. These results contribute to second language research, emphasizing the difficulty of nasal vowels for L2 learners and their implications for language teaching. Overall, nasal vowels pose a consistent challenge, varying based on speakers' native languages.

The current study investigated vowel quality in Miami Cuban Spanish, looking specifically at the influence of English on the Spanish vowel system. The vowel production of eleven Miami Cubans from three generations is investigated. Subjects include six males and five females. Three different elicitation instruments were used. The first was a brief sociolinguistic interview, the second was a story that the participants were asked to read aloud. Carrier words were embedded into the story in order to obtain multiple samples of both stressed and unstressed vowels. For the third instrument, subjects were asked to read a list of words with careful attention to their pronunciation. The reading list contained the same words that were embedded in the story of the first task. These three instruments were repeated in both Spanish and in English in order to investigate possible English influence in the Spanish vowel system of these bilinguals.

The vocal tract shape is crucial to voice production. Its lower part seems particularly relevant for voice timbre. This study analyzes the detailed morphology of parts of the epilaryngeal tube and the hypopharynx for the sustained German vowels /a/, /e/, /i/, /o/, and /u/ by thirteen male singer subjects who were at the beginning of their academic singing studies. Analysis was based on two different phonatory conditions: a natural, speech-like phonation and a singing phonation, like in classical singing. 3D models of the vocal tract were derived from magnetic resonance imaging and compared with long-term average spectrum analysis of audio recordings from the same subjects. Comparison of singing to the speech-like phonation, which served as reference, showed significant adjustments of the lower vocal tract: an average lowering of the larynx by 8 mm and an increase of the hypopharyngeal cross-sectional area (+ 21.9%) and volume (+ 16.8%). Changes in the analyzed epilaryngeal portion of the vocal tract were not significant. Consequently, lower larynx-to-hypopharynx area and volume ratios were found in singing compared to the speech-like phonation. All evaluated measures of the lower vocal tract varied significantly with vowel quality. Acoustically, an increase of high frequency energy in singing correlated with a wider hypopharyngeal area. The findings offer an explanation how classical male singers might succeed in producing a voice timbre with increased high frequency energy, creating a singer‘s formant cluster.

L’objectif de ce travail était d’examiner la spécificité fonctionnelle du traitement et des représentations des hauteurs musicales. À cette fin, ce traitement a été comparé à celui des phonèmes de la parole, d’une part, et aux associations spatiales évoquées par des séquences ordonnées, d’autre part. Nos quatre études avaient pour point commun d’adapter à un nouvel objet de recherche des méthodes bien établies en psychologie cognitive. Ainsi, nous avons exploité la tâche de classification accélérée (Etude 1) de Garner (1974), l’analyse des conjonctions illusoires en mémoire (Etude 2), l’additivité de la composante mismatch negativity (MMN) des potentiels évoqués (Etude 3) et l’observation d’associations spatiales de codes de réponse (Etude 4). Les trois premières études, menées chez des participants non-musiciens, portaient sur la spécificité de traitement des hauteurs par rapport à celui des phonèmes au sein de stimuli chantés. Les deux premières études ont mis en évidence un effet surprenant de la nature des phonèmes sur leurs interactions avec le traitement des mélodies : les voyelles apparaissaient plus intégrées à la mélodie que les consonnes. Ceci était vrai à la fois lors du traitement en temps réel de non-mots chantés (Etude 1) et au niveau des traces en mémoire de ces mêmes non-mots (Etude 2, utilisant une tâche de reconnaissance à choix forcé permettant la mise en évidence de conjonctions illusoires). Cette dissociation entre voyelles et consonnes quant à leur intégration avec les traitements mélodiques ne semblait pas causée par des caractéristiques acoustico-phonétiques telles que la sonorité. Les résultats de la troisième étude indiquaient que les MMNs en réponse à des déviations de hauteur et de voyelle n’étaient pas additives et que leur distribution topographique ne différait pas selon le type de déviation. Ceci suggère que, même au niveau pré-attentionnel, le traitement des voyelles n’est pas indépendant de celui des hauteurs. Dans la quatrième étude, nous avons comparé le traitement des hauteurs musicales à un autre domaine : la cognition spatiale. Nous avons ainsi montré que les non-musiciens comme les musiciens associent les notes graves à la partie inférieure et les notes aiguës à la partie supérieure de l’espace. Les deux groupes liaient aussi les notes graves au côté gauche et les notes aiguës au côté droit, mais ce lien n’était automatique que chez les musiciens. Enfin, des stimuli musicaux plus complexes (intervalles mélodiques) n’évoquaient ces associations spatiales que chez les musiciens et ce, uniquement sur le plan horizontal. Ces recherches contribuent de plusieurs manières à la compréhension de la cognition musicale. Premièrement, nous avons montré que les consonnes et les voyelles diffèrent dans leurs interactions avec la musique, une idée à mettre en perspective avec les rôles différents de ces phonèmes dans l’évolution du langage. Ensuite, les travaux sur les représentations spatiales des hauteurs musicales ouvrent la voie à un courant de recherche qui aidera à dévoiler les liens potentiels entre habiletés musicales et spatiales. / The purpose of this work was to examine the functional specificity of musical pitch processing and representation. To this aim, we compared musical pitch processing to (1) the phonological processing of speech and (2) the spatial associations evoked by ordered sequences. The four studies described here all use classical methods of cognitive psychology, which have been adapted to our research question. We have employed Garner’s (1974) speeded classification task (Study 1), the analysis of illusory conjunctions in memory (Study 2), the additivity of the mismatch negativity (MMN) component of event-related potentials (Study 3), as well as the observation of spatial associations of response codes (Study 4). The three first studies examined, in non-musician participants, the specificity of pitch processing compared to phoneme processing in songs. Studies 1 and 2 revealed a surprising effect of phoneme category on their interactions with melodic processing: vowels were more integrated with melody than were consonants. This was true for both on-line processing of sung nonwords (Study 1) and for the memory traces of these nonwords (Study 2, using a forced-choice recognition task allowing the occurrence of illusory conjunctions). The difference between vowels and consonants was not due to acoustic-phonetic properties such as phoneme sonority. The results of the third study showed that the MMN in response to pitch and to vowel deviations was not additive and that its brain topography did not differ as a function of the kind of deviation. This suggests that vowel processing is not independent from pitch processing, even at the pre-attentive level. In the fourth study, we compared pitch processing to another domain: spatial cognition. We showed that both musicians and non-musicians map pitch onto space, in that they associate low-pitched tones to the lower spatial field and high-pitched tones to the higher spatial field. Both groups of participants also associated low pitched-tones with the left and high-pitched tones with the right, but this association was automatic only in musicians. Finally, more complex musical stimuli such as melodic intervals evoked these spatial associations in the horizontal plane only in musicians. This work contributes to the understanding of music cognition in several ways. First, we have shown that consonants and vowels differ in their interactions with music, an idea related to the contrasting roles of these phonemes in language evolution. Second, the work on the spatial representation of pitch opens the path to research that will help uncover the potential links between musical and spatial abilities.

The articulation of Portuguese nasalized vowels poses some articulatory problems accompanied by negative acoustic effects for the performance of Brazilian art songs. The main objective was to find strategies that permit the singer to conciliate an idiomatic pronunciation of these vowels with a well-balanced resonance, a desirable quality in classical singing. In order to devise these strategies, the author examined sources dealing with nasalized vowels from varied perspectives: acoustic properties of vowel nasalization, phonetic and phonological aspects ofBrazilian Portuguese (BP), historical views on nasality in singing, and recent vocal pedagogy research. In addition to the overall loss of sonority, the main effect of nasalization is felt mainly in the first formant (F1) region of oral vowels, due to the introduction of nasal formants and antiformants, and to shifts in the tongue posture. Several sources report the existence of a nasality contour in BP, by which a nasalized vowel starts with an oral phase and transitions gradually to a nasal phase. The author concludes that the basic approach to sing nasalized vowels in BP is (1) to find the tongue posture corresponding to the oral vowel congener (the “core vowel”), and (2) to adjust the nasality contour in such a way that the oral portion remains prominent in order to keep the resonance balance consistent during the emission of the vowel. Once the core vowel is determined, standard vowel modification choices can be made according to voice type and the musical context in which the vowel is being sung. Some challenging excerpts from Brazilian art songs are examined, with suggestions for the application of the discussed strategies.

S’il est aujourd’hui communément admis que la perception de la parole s’effectue d’une manière plus optimale en mode audiovisuel qu’en mode auditif seul (Benoît, Mohamadi et Kandel, 1994, Schwartz, Berthommier et Savariaux, 2004), la nature des informations visuelles dont il est le plus souvent question est la mimo-gestuelle articulatoire fournie par le locuteur en face à face. Cependant, dans une situation d’enseignement d’une langue étrangère, un autre type d’aide visuelle intervient généralement : la forme écrite des éléments oraux. Pourtant en didactique des langues étrangères, la question du passage à l’écrit est loin d’être consensuelle et certains didacticiens se prononcent en faveur d’un entraînement intensif de la prononciation au tout début de l’apprentissage, avant que l’apprenant ne soit confronté au code écrit (Lauret, 2007). Notre hypothèse est que la dimension facilitante de la forme écrite pour certains publics ne doit pas être négligée, même en début d’apprentissage. Notre recherche se fonde sur des expérimentations menées auprès d’apprenants hindiphones. Tenant compte des spécificités de ce public, nous pensons que l’écrit peut, dans certains cas, faciliter sa réception orale des sons du français en début d’apprentissage, condition nécessaire et préalable à leur production (Renard, 1979). Les tests que nous avons conçus obligent les apprenants à recentrer leur attention sur la graphie de sons (les voyelles nasales [ɑ̃] et [ɔ̃]) dès le début de leur apprentissage au moyen de diverses focalisations visuelles écrites (Fort, Spinelli, Savariaux et Kandel, 2010). Les propositions didactiques que nous formulons à la suite reposent sur la suite logique perception – (transcription graphique) — production même si cette présente étude se centre sur l’évaluation de la perception des sons. / It is commonly admitted today that speech perception is more performing in an audiovisual context than in a visual one (Benoît, Mohamadi and Kandel, 1994, Schwartz, Berthommier and Savariaux, 2004). Visual information in this situation often consists of the speaker’s articulatory and facial gestures provided by the face-to-face interaction. However, when learning a foreign language, another type of visual help is generally available to identify oral forms: their written forms. And yet, in the field of didactics of foreign languages, the issue of the oral-written transition is far from being consensual and some didacticians favour training the pronunciation skills of the learner at the beginning of the learning process, before he is confronted to the written code (Lauret, 2007). Our hypothesis is that the facilitating effect of written forms should not be neglected, even at the beginning of a foreign language learning process. Our research is based on the case of Hindi speakers. Taking into consideration the specificities of this population, we think that written information can, in some cases, facilitate the oral reception of French sounds in the beginning of the learning process, which would be a preliminary condition to their production (Renard, 1979). We have conceived a series of test, forcing the Hindi speaking learners to refocus their attention on the nasal vowels [ɑ̃] and [ɔ̃]’s written form from the beginning of the learning process, using different forms of visual written focuses (Fort, Spinelli, Savariaux and Kandel, 2010). Our didactic proposal relies on the following process: perception – (written form transcription) – production, even though the present study is centred on the evaluation of speech perception.

Esta dissertação objetiva estudar a fala de sergipanos migrantes residentes em São Paulo e em sua região metropolitana, com o intuito de verificar suas taxas de acomodação (TRUDGILL, 1986) à fala paulistana, no que concerne ao grau de abertura das vogais médias pretônicas (como em negócio e coragem), tratado como uma variável numérica (a partir dos valores de F1). Para tanto, os pressupostos teóricos variacionistas (LABOV, 2008 [1972]), o conceito de redes sociais (MILROY, 1987; MILROY & LLAMAS, 2013 [2002]) e algumas definições caras à terceira onda (ECKERT, 2012), como o de identidade (KIESLING, 2013), são tomados como norteadores. O corpus da pesquisa é formado por 27 entrevistas sociolinguísticas realizadas com sergipanos migrantes pertencentes a duas redes sociais distintas. A rede 1 é mais fechada (com uma mobilidade geográfica mais limitada e com um contato menos frequente com paulistanos) e é composta por 16 informantes residentes em Taboão da Serra e Cotia, cidades metropolitanas de São Paulo. A rede 2, por sua vez, é mais aberta (tem mais contato com paulistanos) e é formada por 11 sujeitos, moradores de Osasco, Carapicuíba e São Paulo. Partindo da premissa de que os indivíduos estão se acomodando às pessoas com quem estão conversando rotineiramente, a hipótese central do estudo é que as taxas de acomodação dos migrantes estejam correlacionadas à configuração de suas redes. Dessa maneira, como os sujeitos da rede 2 conversam, com mais frequência, paulistanos, eles tendem a estar mais acomodados à fala paulistana. Além disso, a pesquisa lida com o conceito de identidade com o intuito de testar a hipótese de que o migrante que se identifica mais com Sergipe, do que com São Paulo, tende a estar menos acomodado à nova realidade linguística. Tal teste é realizado com um índice chamado índice de vínculo com Sergipe, criado a partir de um questionário respondido pelos próprios informantes depois da entrevista sociolinguística. O principal resultado das análises é que a hipótese sobre a configuração das redes não se confirma, ou seja, não faz diferença, para as taxas de acomodação, se o informante pertence à rede mais fechada ou à mais aberta. Verificou-se, nesse sentido, que existe uma grande variação entre os indivíduos de uma mesma rede. Ademais, não foi constatada correlação entre o índice de vínculo com Sergipe do migrante e seu grau de acomodação. A análise mais minuciosa de dois migrantes da rede 2 revela que existe uma concentração de vogal /o/ mais baixa quando os falantes estão avaliando as diferentes pronúncias brasileiras e estão discursando sobre sua origem, sobretudo quando usam a palavra nordeste (e suas derivações). Tal descoberta mostra que, em pesquisas sobre acomodação dialetal, é importante voltar-se para a análise do indivíduo e para suas posturas (stance). / This masters thesis presents a study of pre-stressed mid-vowels /e/ (negócio business) and /o/ (coragem courage) in the speech of 27 speakers who migrated from Sergipe to São Paulo. In their original dialect, speakers born and raised in Sergipe (as in other areas of Northeastern Brazil Cardoso, 1999; Callou et al, 2009) tend to open these vowels, relative to their pronunciation by speakers born and raised in São Paulo. Under the assumption that speakers tend to accommodate their speech (TRUDGILL, 1986) to the variety spoken by those with whom they routinely speak, the central hypothesis is that migrants may have different rates of accomodation in the pronunciation of those vowels, depending on the configuration of their networks. That way, migrants whose networks include more Paulistanos would tend to pronounce these vowels more similarly to these speakers, as opposed to migrants whose networks include more Sergipanos and speakers from other Northeastern areas of Brazil. The data were extracted from sociolinguistic interviews, recorded with these migrants according to their participation in two social networks. Network 1 is more closed (limited geographical mobility and less frequent contact with Paulistanos) and comprises 16 migrants living in Taboão da Serra and Cotia, metropolitan cities of São Paulo. Network 2 is more open (more contact with Paulistanos) and comprises 11 subjects, residents of Osasco, Carapicuíba and São Paulo. Data collection and analyses were carried out in accordance with Sociolinguistics (LABOV, 2008 [1972]), the concept of social networks (MILROY, 1987; MILROY & LLAMAS, 2013 [2002]) and third-wave conceptualizations (ECKERT, 2012), such as stance and identity (KIESLING, 2013). Vowel openess was treated as a numeric variable (measured in Praat Boersma & Weenink, 2014 through formant F1, in Hertz). Results do not confirm the network hypothesis, as the degrees of migrants accommodation did not significantly differ according to their network. Rather, there is great variability amongst speakers, including those who are members of the same network. Additionally, there is no correlation between the variable and whether the migrants identify themselves more to the new city or to their Northeastern origins (through an index based on answers to an identity questionnaire performed at the end of the sociolinguistic interviews). Moreover, there is also great variability throughout their interviews. By analyzing more closely the speech of two migrants from Network 2, it is noteworthy that there is a concentration of higher-F1 tokens of /o/ in the word nordeste (Nordeste Northeast), when the speakers are talking about their views on their origins and their original dialect. These findings call for further analysis of individual speech, when it comes to research on dialect accommodation.