Orientador: Odette Gertrudes Luiza Altmann de Souza Campos / Banca: Rosane de Andrade Berlinck / Banca: Ângela Cecília de Souza Rodrigues / Resumo: Este trabalho apresenta um estudo sobre concordância verbal cujo objetivo é o de verificar se o fenômeno em questão se caracteriza como uma regra variável, bem como quais fatores linguísticos e extralinguísticos condicionam essa variação. Para tanto, o escopo do estudo se restringiu à língua falada e escrita pelos moradores da região noroeste do Estado de São Paulo. Utilizaram-se dois corpus para viabilizar a execução da pesquisa, a saber, um pertencente ao IBILCE/ UNESP sob a denominação "O português falado no interior paulista: constituição de um banco de dados anotado para o seu estudo" (PROCESSO FAPESP 03/08058-6), e outro composto por redações escolares de alunos do ensino fundamental I ao ensino médio. Como fatores linguísticos relevantes para nossa análise, destacamos a saliência fônica verbal, o paralelismo formal, a posição do sujeito e a estrutura do sujeito. Como fatores extralinguísticos, a faixa etária e o grau de escolaridade. Utilizamos o programa computacional GOLDVARB-X para realizar a análise quantitativa, cujos resultados mostraram ser a concordância verbal, tanto na escrita quanto na fala, uma regra variável, condicionada principalmente pelo grau de escolaridade dos informantes. Em ambas modalidades, temos 85% de aplicação da regra, porcentagem que consideramos alta / Abstract: This study aims to ascertain whether the verbal agreement is a phenomenon which constitutes a variable rule, and therefore to track the linguistic and extralinguistic factors which would create such a variation. To do so, this scope restricts its scope to the spoken and the written language as it is used by speakers from the northwest of the São Paulo State. Two corpora have been used to make this research feasible, namely, one from IBILCE/UNESP, whose Portuguese title translates as 'The Portuguese spoken in the São Paulo State countryside: the constitution of an annotated database for its study', and another one consisting of pieces of writing by students at either elementary or high school levels. We pay special attention to linguistic factors such as verbal phonetic salience, formal parallelism, the position of the subject as well as to the structure of the subject, which we deem relevant for our analysis. As of extralinguistic factors, we stress the importance of the age group and also of the educational level. We conducted the quantitative analysis using the GOLDVARB-X software, the results of which lead us to figure that the verbal agreement is a variable rule created mainly by the extralinguistic factor concerning the interviewee's level of education, in either the written or in the spoken form. In both situations, the aforementioned rule applies to 85% of cases, a percentage which we deem to be high / Mestre

Com esta pesquisa, tenho como objetivo verificar se é possível estabelecer um diálogo entre o processo de gramaticalização e a reduplicação no português culto, aqui representado pelo material de língua falada sincrônica do século XX (seis inquéritos de D2 - Diálogo entre dois informantes NURC SP CASTILHO e PRETTI, 1987). Entende-se por reduplicação os processos de repetição morfofonológicas, sintáticas e pragmáticas decorrentes, em parte dos casos, de um esquecimento histórico que exige a seleção de um termo com mesmo valor semântico ou função sintática para suprir as necessidades comunicativas. Os dados de reduplicação selecionados do corpus foram organizados segundo quatro padrões funcionais, a saber, REDUPLICAÇÃO por: (i) processamento fônico na palavra; (ii) processamento sintático; (iii) processamento sintático não-regencial; (iv) repetição sintática de orações. Essa organização em quatro padrões funcionais obedeceu tanto às definições utilizadas para reduplicação e redobro encontradas em gramáticas e dicionários especializados (que mencionam desde a repetição de sílabas até a repetição de segmentos maiores) quanto às intuições decorrentes da identificação e da análise de dados encontrados no corpus. A gramaticalização pode ser entendida como transformações que ocorrem na língua e que fazem emergir novos elementos na gramática. É processo de base metafórica, dinâmico e histórico. Nesse processamento histórico, ocorre o deslizamento metafórico em que se pode surpreender o uso de um item por outro, percebido sincronicamente. Avalio os seguintes mecanismos atuantes no processo de gramaticalização e depreendidos das estruturas reduplicadas: o bleaching (SWEETSER, 1988), a reanálise (HEINE et al., 1991), a analogia (HOPPER e TRAUGOTT, 1993), a iconicidade (MARCUSCHI, 2002) e o aumento de freqüência (BYBEE, 2003), bem como procedo à checagem do grau de gramaticalização dos itens analisados, aplicando os parâmetros de Lehmann (1995), os princípios de Hopper (1991) e o continuum de abstratização proposto por Heine, Claudi e Hünnemeyer (1991). / This research has the objective to verify if it is possible to establish a connection between the process of grammaticalization and reduplication in formal Portuguese language. For this verification it was used synchronous spoken language material from the 20th century (six samples of D2 - Dialogue between two informers - NURC SP - CASTILHO and PRETTI, 1987). Reduplication is understood as the processes of morph-phonologic, syntactic and pragmatic repetition resulting from, in some cases, a historical oblivion that claims for the selection of the same semantic value or syntactic function of a word or a term in order to supply the communicative needs. The data of reduplication selected from the corpus had been organized according to four functional patterns, known as: REDUPLICATION by: (i) phonic processing in the word; (II) syntactic processing; (III) non-regencial syntactic processing; (IV) syntactic repetition of statements. This organization in four functional patterns had obeyed the definitions used for reduplication and redoubles found in specialized grammars and dictionaries (in which not only the repetition of syllables but also the repetition of bigger segments are mentioned) as this organization obeyed also the resulting intuitions from the identification and the analysis of the data found in the corpus. The grammaticalization can be understood as transformations that occur in the language and make new elements to emerge in grammar. It is a process of dynamic, historical and metaphoric basis. In this historical processing, the metaphoric sliding occurs in which suddenly it is possible to be surprised with the use of one item by another, and it can be synchronously perceived. The following operating mechanisms were evaluated in the process of grammaticalization which were inferred from the reduplicated structures: bleaching (SWEETSER, 1988), reanalysis (HEINE et al., 1991), analogy (HOPPER and TRAUGOTT, 1993), the iconicity (MARCUSCHI, 2002) and the increase of frequency (BYBEE, 2003). It was also checked the degree of grammaticalization of the analyzed items (samples) throughout the application of Lehmann parameters (1995), the principles of Hopper (1991) and the continuum of abstratitization proposed by Heine, Claudi and Hünnemeyer (1991).

O objetivo do presente trabalho é fazer um estudo sobre as formas de representação da língua falada japonesa nos textos de blogs do gênero idol e as estratégias utilizadas por suas escreventes na sua redação. Para isso, foi utilizado como corpus de análise os dados linguísticos presentes nas postagens desses blogs. Destaque-se que, para a produção de textos em língua japonesa, os escreventes encontram disponíveis quatro sistemas de escrita (o hiragana, o katakana, o kanjie o rmaji), além dos elementos suprassegmentais(sinais de pontuação, diacríticos etc.)que, juntos, oferecem uma variedade de recursos estilísticos e expressivos, os quais são amplamente aproveitados dentro dos textos. Utilizando-se dos blogs idol como cerne, serão feitas discussões acerca da representação da forma oral em seus textos para, em seguida, por meio da análise dos excertos das postagens, descrever os procedimentos linguísticos que resultam em uma língua falada por escrito na língua japonesa. / This work proposes to make a study about the forms of representation found in the Japanese spoken language on idol blog texts and the strategies used in theircomposition. For that purpose, blogs posts will be used as analytical corpus linguistic data. It is noteworthy that the writers have four writing systems at disposal to produce texts in Japanese language (hiragana, katakana, kanji and roomaji), in addition to supra-segmental elements (punctuation marks, diacritics etc.) which together offer a variety of stylistic and expressive resources that are widely used in texts. Using the idol blogs as basis, discussions will be made about the representation of the oral form in it\'s writings by analyzing the excerpts from posts to then describe the linguistic procedures that result in a \"spoken language in writing\" in Japanese.

Syftet med uppsatsen är att studera hur synen på det manliga och kvinnliga talspråket utvecklats från 1900-talet fram till början av 2000-talet och om denna utveckling även syns i elevers ageranden i skolan. De frågeställningar uppsatsen bygger på är: Hur beskriver och förklarar forskare från 2000-talet mäns och kvinnors muntliga språkbruk? Hur beskriver och förklarar forskare från 2000-talet pojkars och flickors muntliga ageranden i skolan? Kan de allmänna teorierna kring mäns och kvinnors muntliga språkbruk relateras till de muntliga ageranden som pojkar och flickor uppvisar i skolan? För att besvara frågeställningarna och leva upp till syftet har en litteraturundersökning genomförts. Forskningen som bearbetas i studien är från 2000-talet och representerar såväl samband mellan språk och kön, som samband mellan språk, kön och skola. Resultatet visar en generell attitydsförändring kring det kvinnliga och manliga talspråket, där de kvinnliga dragen idag värdesätts mer än tidigare. Skillnaderna i mäns och kvinnors talspråk syns även i pojkars och flickors muntliga ageranden i skolan, dock syns inte samma attitydsförändring kring språkbruket i skolan som i samhället. / The purpose of this essay is to study how the attitude to the male and female spoken language has developed during the 20th century and if this development is reflected in school. The essay is built on the following questions: How do researchers of the 21th century describe and explain the spoken language of male and females’? How do researchers of the 21th century describe and explain the spoken language of boys’ and girls’ in school? Can the general theories on male and female spoken language relate to the spoken language of boys and girls in school? To be able to answer these questions a literary study has been carried through. The research that has been processed in the study occurred in the 21th century and represents both the relationship between language and gender as well as the relationship between language, gender and school. The main result shows a general change in attitude on the male and female spoken language, where the female features are more valuable than before. The differences between male and female spoken language can also occur in boys’ and girls’ spoken behavior in school, but the changes in attitude in school haven’t come as far as the changes in attitude in society.

We refactor the traditional two-step approach of automatic speech recognition for spoken language translation. Instead of conventional graphemes, we use phonemes as an intermediate speech representation. Starting with the acoustic model, we revise the cross-lingual transfer and propose a coarse-to-fine method providing further speed-up and performance boost. Further, we review the translation model. We experiment with source and target encoding, boosting the robustness by utilizing the fine-tuning and transfer across ASR and SLT. We empirically document that this conventional setup with an alternative representation not only performs well on standard test sets but also provides robust transcripts and translations on challenging (e.g., non-native) test sets. Notably, our ASR system outperforms commercial ASR systems. 1

The present MA thesis studies adjectives of positive evaluation in present-day spoken British English. Other means of expressing subjective stance (e.g. intensifiers) were repeatedly described as undergoing constant change. However, variation in evaluative adjectives has received little quantitative attention so far (perhaps with the exception of Tagliamonte & Pabst, 2020). Our material consists of two corpora of informal spoken British English: Spoken BNC2014 and the spoken, demographically sampled section of the original BNC (1994). The starting point for the analysis is a frequency list of adjectives, from which we select adjectives with evaluative potential which differ significantly in frequency across the two corpora. Three adjectives: amazing, awesome and cool, are described in greater detail. The use of evaluative adjectives is described from several perspectives. We focus on syntactic functions of the adjectives, their co-occurrence with intensifiers and their collocations. From a sociolinguistic point of view, we describe the use of the adjectives with respect to the age and gender of the speakers. The analysis showed that adjectives of positive evaluation are an unstable category, and the change in adjectival use was reflected in its distribution. At the same time, we note changes in...

京都大学 / 0048 / 新制・課程博士 / 博士(人間・環境学) / 甲第19808号 / 人博第779号 / 新制||人||187(附属図書館) / 27||人博||779(吉田南総合図書館) / 32844 / 京都大学大学院人間・環境学研究科共生人間学専攻 / (主査)教授 東郷 雄二, 教授 藤田 耕司, 教授 田窪 行則 / 学位規則第4条第1項該当 / Doctor of Human and Environmental Studies / Kyoto University / DFAM

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第24256号 / 情博第800号 / 新制||情||135(附属図書館) / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 河原 達也, 教授 森 信介, 教授 伊藤 孝行 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

BACKGROUND: Many families of children who are deaf or hard of hearing (DHH) pursue the development of spoken language through the use of advanced hearing technology and intensive, specialized listening and spoken language (LSL) intervention services. LSL practitioners utilize caregiver coaching to transfer knowledge and skills to parents, equipping them to effectively support their child's listening and language development. Caregiver coaching builds parents' capacity to implement intervention strategies within daily routines to maximize learning opportunities to reach this goal. While caregiver coaching is a hallmark of LSL practice, there is a lack of consensus and paucity of evidence to support its use with children who are DHH learning to listen and talk. The purpose of this research was therefore to gain a greater understanding of caregiver coaching practices in LSL services for families of children ages birth-3 who are DHH. OBJECTIVES: 1) to provide a synthesis of the literature and identify gaps in the existing knowledge base regarding coaching in LSL services, 2) to gain insight into practitioners' and caregivers' experiences with coaching in LSL services, and 3) to gain a better understanding of coaching practices in LSL services with families of children who are DHH. METHODOLOGY: The first objective was addressed by performing a scoping review of the literature to synthesize the relevant research and professional practice recommendations and identify gaps in the knowledge base regarding coaching caregivers in LSL practice. The remaining objectives were addressed through qualitative, semi-structured interviews and video observation discussions with LSL practitioners and caregivers to gain insight into caregivers' perspectives and practitioners' perspectives and practices. Interviews were conducted with 14 practitioners and 13 caregivers at three intervention sites in the US and Canada. These sites represent three different service delivery models of LSL intervention for families of children who are DHH. RESULTS: The scoping review findings revealed a lack of consensus in the literature regarding the principles and practices of caregiver coaching in LSL services. We presented the following topics found in the literature: coaching practices, training for coaching, the effectiveness of coaching, and recommendations for coaching. Caregivers reported coaching as a positive experience; however, coaching practices differed among sites and between practitioners, supporting the results of the scoping review. The caregivers indicated factors that contribute to a positive coaching relationship, including practitioner characteristics, establishing explicit expectations, and adapting to caregivers' changing needs over time. The practitioner interviews also supported a lack of consistency in coaching practices between sites and indicated that underlying beliefs impact how practitioners coach and engage caregivers. CONCLUSION: The findings from this dissertation provide practical, actionable steps that LSL practitioners can implement to develop and support the caregiver coaching relationship. These findings have the potential to inform professional preparation and development activities to better equip practitioners to engage caregivers in the intervention process, and ultimately, positively impact the listening and spoken language outcomes of the children they serve.

Sequence-to-sequence learning led to significant improvements to machine translation (MT) and automatic speech recognition (ASR) systems. These advancements were first reflected in spoken language translation (SLT) when using a cascade of (at least) ASR and MT with the new "neural" models, then by using sequence-to-sequence learning to directly translate the input audio speech into text in the target language. In this thesis we cover both approaches to the SLT task. First, we show the limits of NMT in terms of robustness to input errors when compared to the previous phrase-based state of the art. We then focus on the NMT component to achieve better translation quality with higher computational efficiency by using a network based on weakly-recurrent units. Our last work involving a cascade explores the effects on the NMT robustness when adding automatic transcripts to the training data. In order to move to the direct speech-to-text approach, we introduce MuST-C, the largest multilingual SLT corpus for training direct translation systems. MuST-C increases significantly the size of publicly available data for this task as well as their language coverage. With such availability of data, we adapted the Transformer architecture to the SLT task for its computational efficiency . Our adaptation, which we call S-Transformer, is meant to better model the audio input, and with it we set a new state of the art for MuST-C. Building on these positive results, we finally use S-Transformer with different data applications: i) one-to-many multilingual translation by training it on MuST-C; ii participation to the IWSLT 19 shared task with data augmentation; and iii) instance-based adaptation for using the training data at test time. The results in this thesis show a steady quality improvement in direct SLT. Our hope is that the presented resources and technological solutions will increase its adoption in the near future, so to make multilingual information access easier in a globalized world.