Este trabalho apresenta um estudo sobre a utilização de dispositivos adaptativos para realizar tradução texto-voz. O foco do trabalho é a criação de um método para a tradução grafema-fonema para a língua portuguesa baseado em autômatos adaptativos e seu uso em um software de tradução texto-voz. O método apresentado busca mimetizar o comportamento humano no tratamento de regras de tonicidade, separação de sílabas e as influências que as sílabas exercem sobre suas vizinhas. Essa característica torna o método facilmente utilizável para outras variações da língua portuguesa, considerando que essas características são invariantes em relação à localidade e a época da variedade escolhida. A variação contemporânea da língua falada na cidade de São Paulo foi escolhida como alvo de análise e testes neste trabalho. Para essa variação, o modelo apresenta resultados satisfatórios superando 95% de acerto na tradução grafema-fonema de palavras, chegando a 90% de acerto levando em consideração a resolução de dúvidas geradas por palavras que podem possuir duas representações sonoras e gerando uma saída sonora inteligível aos nativos da língua por meio da síntese por concatenação baseada em sílabas. Como resultado do trabalho, além do modelo para tradução grafema-fonema de palavras baseado em autômatos adaptativos, foi criado um método para escolha da representação fonética correta em caso de ambigüidade e foram criados dois softwares, um para simulação de autômatos adaptativos e outro para a tradução grafema-fonema de palavras utilizando o modelo de tradução criado e o método de escolha da representação correta. Esse último software foi unificado ao sintetizador desenvolvido por Koike et al. (2007) para a criação de um tradutor texto-voz para a língua portuguesa. O trabalho mostra a viabilidade da utilização de autômatos adaptativos como base ou como um elemento auxiliar para o processo de tradução texto-voz na língua portuguesa. / This work presents a study on the use of adaptive devices for text-to-speech translation. The work focuses on the development of a grapheme-phoneme translation method for Portuguese based on Adaptive Automata and the use of this method in a text-to-speech translation software. The presented method resembles human behavior when handling syllable separation rules, syllable stress definition and influences syllables have on each other. This feature makes the method easy to use with different variations of Portuguese, since these characteristics are invariants of the language. Portuguese spoken nowadays in São Paulo, Brazil has been chosen as the target for analysis and tests in this work. The method has good results for such variation of Portuguese, reaching 95% accuracy rate for grapheme-phoneme translation, clearing the 90% mark after resolution of ambiguous cases in which different representations are accepted for a grapheme and generating phonetic output intelligible for native speakers based on concatenation synthesis using syllables as concatenation units. As final results of this work, a model is presented for grapheme-phoneme translation for Portuguese words based on Adaptive Automata, a methodology to choose the correct phonetic representation for the grapheme in ambiguous cases, a software for Adaptive Automata simulation and a software for grapheme-phoneme translation of texts using both the model of translation and methodology for disambiguation. The latter software was unified with the speech synthesizer developed by Koike et al. (2007) to create a text-to-speech translator for Portuguese. This work evidences the feasibility of text-to-speech translation for Portuguese using Adaptive Automata as the main instrument for such task.

Esta tese objetiva analisar o desvozeamento variável das plosivas bilabial, alveolar e velar (abacaxi~apacaxi, dela~tela e Glória~Clória) do português brasileiro em contato com o hunsrückisch, língua de imigração alemã. A variedade de português investigada é a falada em Glória, comunidade da zona rural do município de Estrela, Rio Grande do Sul, Brasil. O status social da variável em estudo é estereotipado e as atitudes linguísticas são analisadas em relação ao desvozeamento das plosivas. Investiga-se, portanto, a relação entre o processo variável e as atitudes linguísticas dos falantes para com o português brasileiro local e com a língua de imigração. O estudo orienta-se pela Sociolinguística Variacionista (LABOV, 2008 [1972]) e pelo estudo de atitudes linguísticas (TRIANDIS, 1974; FASOLD, 1996; KAUFMANN, 1997; 2011; GILES e BILLINGS, 2004; GARRET, 2005; VANDERMEEREN, 2005; LABOV, 2010). Para a análise de regra variável, foram levantados contextos de desvozeamento de vinte e quatro entrevistas sociolinguísticas de informantes de Glória. Os dados foram submetidos à análise estatística pelo pacote computacional VARBRUL, versão GoldVarb X, para verificar os fatores linguísticos e extralinguísticos que condicionam o desvozeamento variável das plosivas. Constatou-se que a proporção de desvozeamento é baixa, 2,6%. Os informantes do gênero feminino, com menor grau de escolarização, ensino fundamental, e com mais de 47 anos condicionam o processo. As palavras com maior número de sílabas, contexto precedente vazio e o contexto seguinte alveolar, sílabas pretônica e tônica favorecem o desvozeamento das plosivas. Para o estudo de atitudes linguísticas, foi realizada uma pesquisa qualitativa mediante a aplicação do questionário “As atitudes linguísticas no português brasileiro em contato com o hunsrückisch”, adaptado de Kaufmann (1997; 2011). Posteriormente, os dados levantados com o questionário foram submetidos a tratamento estatístico pelo software IBM SPSS, versão 22.0. Verificou-se que os núcleos familiares, compostos por avós, pais, irmãos, tios e tias, influenciam as práticas sociais, linguísticas e culturais na comunidade de Glória, principalmente a figura feminina, a mãe, (geração mais velha) que tem responsabilidade na formação e preservação cultural nas antigas áreas de imigração europeia no sul do Brasil e também pela função social que exerce em casa, na educação inicial dos filhos. Os informantes mais jovens realizam mais atividades de trabalho, diversão e lazer do que seus pais, usando o português como língua de interação. Tais atividades influenciam as atitudes dos falantes em relação à preferência a usar português em lugar do hunsrückisch. Os resultados evidenciam que a escolaridade também contribui para as atitudes, com práticas linguísticas em português brasileiro local no ambiente escolar: os falantes orientam-se à cultura brasileira e à fala em português brasileiro local e buscam fugir ao estereótipo do desvozeamento das plosivas. / This thesis aims at analyzing the variable devoicing of bilabial, alveolar and velar plosives (abacaxi~apacaxi, dela~tela e Glória~Clória) in Brazilian Portuguese in contact with hunsrückisch, German immigration language. The Portuguese variety under investigation is the one spoken in Glória, a rural area community in the town of Estrela, Rio Grande do Sul, Brazil. The social status of the variable in study is stereotyped and the linguistic attitudes are analyzed in relation to the devoicing of the plosives. It investigated, therefore, the relation between the variable process and the speakers’ linguistic attitudes towards local Brazilian Portuguese and the immigration language. The study is oriented by Variationist Sociolinguistics (LABOV, 2008 [1972]) and by linguistic attitude studies (TRIANDIS, 1974; FASOLD, 1996; KAUFMANN, 1997; 2011; GILES e BILLINGS, 2004; GARRET, 2005; VANDERMEEREN, 2005; LABOV, 2010). For analyzing the variable rule, devoicing contexts, present in twenty-four sociolinguistic interviews with informants from Glória, were gathered. The data were submitted to statistical analysis by the computational package VARBRUL, version GoldVarb X, in order to verify the linguistic and extra-linguistic factors that condition the variable devoicing of plosives. It was determined that the devoicing proportion is low, 2.6%. The female gender informants, with the lowest scholarization degree, middle school, and older than 47 years old condition the process. The words with a bigger number of syllables, empty precedent context and alveolar posterior context, pre-tonic and tonic syllables favor the plosives devoicing. For the study of linguistic attitudes, a qualitative research was carried out, through the application of the questionnaire “As atitudes linguísticas no português brasileiro em contato com o hunsrückisch”, adapted from Kaufmann (1997; 2011). After that, the data collected through the questionnaire were submitted to statistical treatment by the software IBM SPSS, version 22.0. It was possible to verify that family cores, composed by grandparents, parents, siblings, uncles and aunts, influence the social, linguistic and cultural practices in the community of Glória, specially the female figure, the mother, (older generation) which holds responsibility in cultural formation and maintenance in old European immigration areas in the south of Brazil and also in the social function exerted at home, in the initial education of the children. The younger informants perform more activities related to work, entertainment and leisure than their parents, using the Portuguese as interaction language. Such activities influence the speakers’ attitudes towards the preference for using Portuguese over hunsrückisch. The results show that scholarization also contributes for the attitudes, with linguistic practices in local Brazilian Portuguese in the school environment: the speakers orient themselves towards the Brazilian culture and the speech in local Brazilian Portuguese, and seek to scape from the stereotype of plosives devoicing.

Cette mémoire a pour objectif de mettre en évidence le délaissement dont a souffert et souffre encore dans l’actualité l’enseignement de la prononciation dans les cours d’espagnol comme langue étrangère (ELE), réaffirmer son importance et défendre la nécessité de son intégration dans les salles de classe. C’est un fait qu’avec l’actuelle approche communicative et « par devoirs », les étudiants acquièrent des connaissances supérieures en langue écrite et des résultats peu satisfaisants dans la prononciation. D’autre part, les enseignants ne comptent pas sur des méthodes d’intégration de la prononciation de ces approches, et la carence de matériels se fait plus qu’évidente. Les conceptions curriculaires actuelles n’ont pas intégré de forme naturelle l’enseignement de la prononciation dans ses exposés méthodologiques. La sélection et analyse de certains des manuels d’enseignement d’ELE, avec ce dont on travaille actuellement dans la province du Québec, confirme ces carences. Dans la dernière partie du travail on présente une sélection d’exercices que l’on considère utiles pour l’enseignement et le développement de la prononciation, incluant les éléments suprasegmentaires, d’une façon ludique, dans un contexte réel et communicatif et de manière intégrée. Une didactique de l’enseignement qui prête attention au système formel de la langue, la grammaire et le lexique, peut aussi contempler l’enseignement de la prononciation depuis la même perspective. L’important c’est qu’elle soit intégrée comme une phase supplémentaire du processus d’apprentissage, et cesse d’être une matière à repasser dans les curriculums. / This memory aims to highlight the renunciation that suffered, and still do nowadays, the teaching of the pronunciation in the Spanish courses as foreign language (SFL), to reaffirm its importance and to defend the need for its integration in the classrooms. It is a fact that with the current communicative approach and with the different duties, the students acquire higher knowledge in written language but do not have very satisfactory results in their pronunciation. In addition, the teachers don’t count on the integration of pronunciation methods in their approaches, and the deficiency of this material is more than obvious. The current curricular designs did not integrate a natural form of pronunciation teaching in its methodological talks. The selection and analyzes of some IT teaching handbooks, that are currently use in the province of Quebec, confirms these deficiencies. In the last part of this work, we present a selection of exercises that we consider useful for the teaching and the development of pronunciation, including the elements supra segmentary in a real and communicative context and in an integrated way. Didactic of the teaching that pays attention to the formal language system, grammar and the lexicon, can also contemplate pronunciation teaching by using the same prospect. Important it is that this perspective should be integrated like an additional phase of the training process, and stop to be a matter to be passed by in the curriculums.

Made available in DSpace on 2016-04-28T18:22:11Z (GMT). No. of bitstreams: 1 Jae Keum Oh.pdf: 6453251 bytes, checksum: 44935e77974596a4e82a9c7b7c032ce6 (MD5) Previous issue date: 2010-10-29 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / This project explores the issues of expressivity in speech by analyzing the nature of phonetic-acoustic and perceptive. The goal is to investigate the effects of pauses in the interpretation of four poems from the four main heteronyms of Fernando Pessoa by an announcer-actor. The object of this investigation is the relationship established between sound and meaning in this interpretation. The corpora are composed of a poem by each of the four main heteronyms Fernando Pessoa's Alberto Caeiro, Ricardo Reis, Álvaro de Campos and Fernando Pessoa by himself, recorded on CD. Acoustic Phonetic analysis includes the measurement of the pause using the software Praat. The results show the use of pauses by the speaker-actor according to the poetic text or backronym. The pauses in poetry reading have a role as developer of the architecture of oral interpretation or expression of meaning effects / Este projeto explora a temática da Expressividade na fala por meio de análise de natureza fonético-acústica e perceptiva. O objetivo é investigar os efeitos das pausas na interpretação de quatro poemas dentre os quatro principais heterônimos de Fernando Pessoa por um locutor-ator. O objeto desta investigação é a relação que se estabelece entre som e sentido nessa interpretação. Os corpora são compostos por um poema de cada um dos quatro principais heterônimos de Fernando Pessoa: Alberto Caeiro, Ricardo Reis, Álvaro de Campos e Fernando Pessoa por ele mesmo, gravados em CD. A análise Fonético Acústica compreende a medição da pausa por meio do software Praat. Os resultados mostram o uso das pausas pelo locutor-ator em função do texto poético ou heteronímico. As pausas, em leitura de poesia, têm o papel de colaborador da arquitetura da interpretação oral ou de expressão dos efeitos de sentidos

The modern computerised era in which we live means that all areas of society are subjected to information technology. Children are increasingly exposed to and have access to computers, which necessarily have an influence on their education and development. Seeing that children are the most receptive group when it comes to exposure to new technology, this may have enriching effects on computer literacy, which is often considered part of general literacy. With regards to literacy, reading is one of the skills which pose a high challenge. It is also a key skill that opens the world of information, as most information is received through reading. Any reading stimulation via electronic or printed media that encourages learners to read is considered to improve reading ability, reading speed, reading comprehension and phonetic awareness. The aim of this study was to determine whether there was a correlation between early computer exposure and experience (both computer access and use) and reading ability of grade 1 learners. In this study a one shot cross-sectional survey design was used. Schools with grade 1 learners in the Umvoti district of Kwa Zulu Natal were selected to participate in the study - stratified sampling was used. The grade 1 classes were selected on the basis of random sampling. The selected classes (N=4; 85 learners) participated 'intact' in the study. The data was analysed using descriptive as well as inferential statistics (e.g. MANCOVA). / Thesis (M.Ed.)--North-West University, Potchefstroom Campus, 2007.

Cette mémoire a pour objectif de mettre en évidence le délaissement dont a souffert et souffre encore dans l’actualité l’enseignement de la prononciation dans les cours d’espagnol comme langue étrangère (ELE), réaffirmer son importance et défendre la nécessité de son intégration dans les salles de classe. C’est un fait qu’avec l’actuelle approche communicative et « par devoirs », les étudiants acquièrent des connaissances supérieures en langue écrite et des résultats peu satisfaisants dans la prononciation. D’autre part, les enseignants ne comptent pas sur des méthodes d’intégration de la prononciation de ces approches, et la carence de matériels se fait plus qu’évidente. Les conceptions curriculaires actuelles n’ont pas intégré de forme naturelle l’enseignement de la prononciation dans ses exposés méthodologiques. La sélection et analyse de certains des manuels d’enseignement d’ELE, avec ce dont on travaille actuellement dans la province du Québec, confirme ces carences. Dans la dernière partie du travail on présente une sélection d’exercices que l’on considère utiles pour l’enseignement et le développement de la prononciation, incluant les éléments suprasegmentaires, d’une façon ludique, dans un contexte réel et communicatif et de manière intégrée. Une didactique de l’enseignement qui prête attention au système formel de la langue, la grammaire et le lexique, peut aussi contempler l’enseignement de la prononciation depuis la même perspective. L’important c’est qu’elle soit intégrée comme une phase supplémentaire du processus d’apprentissage, et cesse d’être une matière à repasser dans les curriculums. / This memory aims to highlight the renunciation that suffered, and still do nowadays, the teaching of the pronunciation in the Spanish courses as foreign language (SFL), to reaffirm its importance and to defend the need for its integration in the classrooms. It is a fact that with the current communicative approach and with the different duties, the students acquire higher knowledge in written language but do not have very satisfactory results in their pronunciation. In addition, the teachers don’t count on the integration of pronunciation methods in their approaches, and the deficiency of this material is more than obvious. The current curricular designs did not integrate a natural form of pronunciation teaching in its methodological talks. The selection and analyzes of some IT teaching handbooks, that are currently use in the province of Quebec, confirms these deficiencies. In the last part of this work, we present a selection of exercises that we consider useful for the teaching and the development of pronunciation, including the elements supra segmentary in a real and communicative context and in an integrated way. Didactic of the teaching that pays attention to the formal language system, grammar and the lexicon, can also contemplate pronunciation teaching by using the same prospect. Important it is that this perspective should be integrated like an additional phase of the training process, and stop to be a matter to be passed by in the curriculums.

Approximate string matching consists in identifying strings as similar even ifthere is a number of mismatch between them. This technique is one of thesolutions to reduce the exact matching strictness in data comparison. In manycases it is useful to identify stream variation (e.g. audio) or word declension (e.g.prefix, suffix, plural). Approximate string matching can be used to score terms in InformationRetrieval (IR) systems. The benefit is to return results even if query terms doesnot exactly match indexed terms. However, as approximate string matchingalgorithms only consider characters (nor context neither meaning), there is noguarantee that additional matches are relevant matches. This paper presents the effects of some approximate string matchingalgorithms on search results in IR systems. An experimental research design hasbeen conducting to evaluate such effects from two perspectives. First, resultrelevance is analysed with precision and recall. Second, performance is measuredthanks to the execution time required to compute matches. Six approximate string matching algorithms are studied. Levenshtein andDamerau-Levenshtein computes edit distance between two terms. Soundex andMetaphone index terms based on their pronunciation. Jaccard similarity calculatesthe overlap coefficient between two strings. Tests are performed through IR scenarios regarding to different context,information need and search query designed to query on a technicaldocumentation related to software development (man pages from Ubuntu). Apurposive sample is selected to assess document relevance to IR scenarios andcompute IR metrics (precision, recall, F-Measure). Experiments reveal that all tested approximate matching methods increaserecall on average, but, except Metaphone, they also decrease precision. Soundexand Jaccard Similarity are not advised because they fail on too many IR scenarios.Highest recall is obtained by edit distance algorithms that are also the most timeconsuming. Because Levenshtein-Damerau has no significant improvementcompared to Levenshtein but costs much more time, the last one is recommendedfor use with a specialised documentation. Finally some other related recommendations are given to practitioners toimplement IR systems on technical documentation.

See file

The modern computerised era in which we live means that all areas of society are subjected to information technology. Children are increasingly exposed to and have access to computers, which necessarily have an influence on their education and development. Seeing that children are the most receptive group when it comes to exposure to new technology, this may have enriching effects on computer literacy, which is often considered part of general literacy. With regards to literacy, reading is one of the skills which pose a high challenge. It is also a key skill that opens the world of information, as most information is received through reading. Any reading stimulation via electronic or printed media that encourages learners to read is considered to improve reading ability, reading speed, reading comprehension and phonetic awareness. The aim of this study was to determine whether there was a correlation between early computer exposure and experience (both computer access and use) and reading ability of grade 1 learners. In this study a one shot cross-sectional survey design was used. Schools with grade 1 learners in the Umvoti district of Kwa Zulu Natal were selected to participate in the study - stratified sampling was used. The grade 1 classes were selected on the basis of random sampling. The selected classes (N=4; 85 learners) participated 'intact' in the study. The data was analysed using descriptive as well as inferential statistics (e.g. MANCOVA). / Thesis (M.Ed.)--North-West University, Potchefstroom Campus, 2007.

Stops are one of the most common sounds across languages of the world. Among these pervasive sounds, ejectives form a unique group that is distinguishable from other types of stops. Their particular mechanism of articulation, such as larynx raising and unusually high oral pressure, separates them from the others. More interestingly, a listener perceives them differently and makes a distinction from non-ejective, or pulmonic, stops. What is it that we perceive when hearing ejectives? Do we perceive certain acoustic cues or auditory qualities that are part of their distinctive phonetic nature? Are these phonetic characteristics always distinctive? In other words, is our perception of the ejectives always consistent without any variation at the phonetic level? Motivated by these questions and from my recent exposure to Tsilhqut’in ejectives, I set out to pursue a phonetic investigation of these intriguing sounds. The present study is composed of two main analyses. One is an acoustic analysis that instrumentally examines a dataset of ejective and non-ejective stops in the Tsilhqut’in language with respect to acoustic dimensions such as Voice Onset Time (VOT) in order to compare all the stop classes in terms of their acoustic properties. Such a comparison helps to phonetically characterize the ejectives within the language. The acoustic measures also enable us to compare the characteristics of Tsilhqut’in ejectives with those in other languages, based on previously reported acoustic correlates. In order to determine the characteristics of ejectives across languages, Tsilhqut’in ejectives were compared with ejectives in different languages (e.g., Inguish). The other analysis is auditory, whereby I have examined how I perceived a subset of the ejectives taken out of the whole dataset and compared my auditory judgments with the acoustic measurements in order to find any correlation between results from the two analyses. The findings of the study indicate that Tsilhqut’in ejectives do not follow a traditional binary typology of ejectives. That is, they are neither strong nor weak, as is often claimed in the literature. They are congruent with what recent studies (e.g., Warner 1996) have found of ejectives in other languages – phonetic variability. This means that the dichotomy cannot account for the variability in ejectives at the phonetic level and that an optimal way of classifying ejectives across languages still awaits discovery. To the best of my knowledge, no other phonetic study has been conducted on Tsilhqut’in ejectives prior to the current study. Moreover, there has been little research or documentation carried out on any other phonetic aspects or sounds of this Athabaskan language. I expect that this instrumental study will contribute to the field of linguistics by adding new phonetic knowledge about such a rarely studied language, and I also expect the present study to play a role in the understanding of language learning and of language revitalization around the world.