The rationale for this study was to investigate the implementation of the THRASS literacy programme on a deaf learner who uses the spoken language approach. Particular emphasis was given to the role played by the Phoneme Machine together with Cued Speech. THRASS focuses on phoneme-grapheme correspondence by explicit phonics instruction to develop word analysis and recognition skills. Cued Speech is used as an instructional tool to facilitate visual access to auditory-based phonology. The research was framed within the Interpretivist paradigm and a qualitative case study design predominated, although the launch and landing of the study was quantitative in nature. The findings indicated that the auditory-based phonology of the English language may be accessed by a deaf learner, when supported by a visual instructional tool such as Cued Speech in synchronicity with speech-reading, to develop print literacy skills. This study opens the gateway to further enquiry on enhancing deaf literacy levels. / Inclusive Education / M. Ed. (Inclusive Education)

Cette thèse s'inscrit dans le cadre de l'émergence de l'apprentissage profond et aborde la compréhension de la parole assimilée à l'extraction et à la représentation automatique du sens contenu dans les mots d'une phrase parlée. Nous étudions une tâche d'étiquetage en concepts sémantiques dans un contexte de dialogue oral évaluée sur le corpus français MEDIA. Depuis une dizaine d'années, les modèles neuronaux prennent l'ascendant dans de nombreuses tâches de traitement du langage naturel grâce à des avancées algorithmiques ou à la mise à disposition d'outils de calcul puissants comme les processeurs graphiques. De nombreux obstacles rendent la compréhension complexe, comme l'interprétation difficile des transcriptions automatiques de la parole étant donné que de nombreuses erreurs sont introduites par le processus de reconnaissance automatique en amont du module de compréhension. Nous présentons un état de l'art décrivant la compréhension de la parole puis les méthodes d'apprentissage automatique supervisé pour la résoudre en commençant par des systèmes classiques pour finir avec des techniques d'apprentissage profond. Les contributions sont ensuite exposées suivant trois axes. Premièrement, nous développons une architecture neuronale efficace consistant en un réseau récurent bidirectionnel encodeur-décodeur avec mécanisme d’attention. Puis nous abordons la gestion des erreurs de reconnaissance automatique et des solutions pour limiter leur impact sur nos performances. Enfin, nous envisageons une désambiguïsation de la tâche de compréhension permettant de rendre notre système plus performant. / This thesis is a part of the emergence of deep learning and focuses on spoken language understanding assimilated to the automatic extraction and representation of the meaning supported by the words in a spoken utterance. We study a semantic concept tagging task used in a spoken dialogue system and evaluated with the French corpus MEDIA. For the past decade, neural models have emerged in many natural language processing tasks through algorithmic advances or powerful computing tools such as graphics processors. Many obstacles make the understanding task complex, such as the difficult interpretation of automatic speech transcriptions, as many errors are introduced by the automatic recognition process upstream of the comprehension module. We present a state of the art describing spoken language understanding and then supervised automatic learning methods to solve it, starting with classical systems and finishing with deep learning techniques. The contributions are then presented along three axes. First, we develop an efficient neural architecture consisting of a bidirectional recurrent network encoder-decoder with attention mechanism. Then we study the management of automatic recognition errors and solutions to limit their impact on our performances. Finally, we envisage a disambiguation of the comprehension task making the systems more efficient.

The overall theme of this diploma thesis are homilies entitled to Saint Anna by two baroque homilists, who origin from various language backgrounds. This thesis is divided into two major sections. The first section includes general information about the important baroque preachers {--} with a focus on Bílovský and de Waldt. It also provides valuable information about the contemporary grammars. The first section of this thesis is primarily based on a proven literature theory. The second section of this thesis includes practical analysis of three specific homilies on the phonological and morphological levels. The analysis is based on the theme, style, and lexicon pages of content as well as the confrontation of their means of expression. The objective of this thesis is a comparison of the languages, shown by both preachers through phonology, morphology, and the judgment of a level to which spoken language extends into their language.

In deze scriptie wordt onderzocht hoe spreektaal in de Nederlandse kinderboeken Krassen in het tafelblad en Ik ben Polleke hoor! vertaald is naar het Zweeds. De analyse is gemaakt volgens de theorie over vertaalnormen van de descriptieve vertaalwetenschap, en uitdrukkingen van spreektaal worden geanalyseerd vanuit Lindqvists (2005) indeling op drie niveaus: fonologisch/ morfologisch niveau, lexicaal niveau en syntactisch niveau. Het onderzoek toont aan welke strategie de vertalers mee hebben gewerkt en de vertaalnormen die mogelijk invloed hebben gehad tijdens het vertalen. Het resultaat wijst op dat de spreektaal in Krassen in het tafelblad vrijer vertaald is dan de spreektaal in Ik ben Polleke hoor!. Beide vertalingen tonen een streven aan om binnen de doelcultuur te passen, maar Krassen in het tafelblad ligt dichter bij een aanvaardbare vertaling dan Ik ben Polleke hoor!, die zich dichter bij de brontekst bevindt en gezien kan worden als een meer adequate vertaling. / This thesis examines how spoken language in the Dutch children's books Krassen in het tafelblad and Ik ben Polleke hoor! has been translated into Swedish. The analysis is done according to the descriptive translation studies’ theory of translation norms, and spoken language expressions are analyzed by using Lindqvist's (2005) classification of spoken language markers on three levels: the phonological/ morphological level, the lexical level and the syntactic level. The survey shows the translation strategies used by the translators and the norms that may have had an impact during the translation process. The result shows that the spoken language in Krassen in het tafelblad is more freely translated than the spoken language in the translation of Ik ben Polleke hoor!. Both translations show an ambition to fit within their target culture, but Krassen in het tafelblad lies closer to an acceptance-oriented translation than Ik ben Polleke hoor!, that adheres more to the source text and can be seen as a more adequate-oriented translation.

The rationale for this study was to investigate the implementation of the THRASS literacy programme on a deaf learner who uses the spoken language approach. Particular emphasis was given to the role played by the Phoneme Machine together with Cued Speech. THRASS focuses on phoneme-grapheme correspondence by explicit phonics instruction to develop word analysis and recognition skills. Cued Speech is used as an instructional tool to facilitate visual access to auditory-based phonology. The research was framed within the Interpretivist paradigm and a qualitative case study design predominated, although the launch and landing of the study was quantitative in nature. The findings indicated that the auditory-based phonology of the English language may be accessed by a deaf learner, when supported by a visual instructional tool such as Cued Speech in synchronicity with speech-reading, to develop print literacy skills. This study opens the gateway to further enquiry on enhancing deaf literacy levels. / Inclusive Education / M. Ed. (Inclusive Education)

Studien undersöker hur de engelska slanguttrycken och talspråksmarkörerna i ungdomsromanen Paper Towns av John Green översatts till nederländska och svenska. En analys av slanguttrycken och talspråksmarkörerna i romanens första kapitel, utifrån Lambert & van Gorps modell (1985), visar vilka översättningsstrategier som översättarna valt och vilka översättningsnormer dessa är ett uttryck för. Resultatet visar att översättningarnas preliminära data pekar mot en adekvansinriktad översättningsstrategi. På mikronivå visar resultatet en mer acceptansinriktad översättningsstrategi vid översättning av slang och talspråksmarkörer, med en något större källspråksinriktning i den nederländska översättningen. / The study investigates how slang and spoken language markers in English in the Young Adult novel Paper Towns by John Green have been translated into Dutch and Swedish. An analyse of the expressions found in the first chapter of the novel, based on the method created by Lambert and van Gorp (1985), shows the translation strategies and the underlying translational norms. The results of the analysis of the preliminary data of the translations points towards adequate translation strategies. At a micro level, the results indicate more acceptable translation strategies regarding the translation of slang and spoken language markers, with slightly more source-orientation in the Dutch translation.

South Africa has eleven official languages, ten of which are considered “resource-scarce”. For these languages, even basic linguistic resources required for the development of speech technology systems can be difficult or impossible to obtain. In this thesis, the process of developing Spoken Language Identification (S-LID) systems in resource-scarce environments is investigated. A Parallel Phoneme Recognition followed by Language Modeling (PPR-LM) architecture is utilized and three specific scenarios are investigated: (1) incomplete resources, including the lack of audio transcriptions and/or pronunciation dictionaries; (2) inconsistent resources, including the use of speech corpora that are unmatched with regard to domain or channel characteristics; and (3) poor quality resources, such as wrongly labeled or poorly transcribed data. Each situation is analysed, techniques defined to mitigate the effect of limited or poor quality resources, and the effectiveness of these techniques evaluated experimentally. Techniques evaluated include the development of orthographic tokenizers, bootstrapping of transcriptions, filtering of low quality audio, diarization and channel normalization techniques, and the human verification of miss-classified utterances. The knowledge gained from this research is used to develop the first S-LID system able to distinguish between all South African languages. The system performs well, able to differentiate among the eleven languages with an accuracy of above 67%, and among the six primary South African language families with an accuracy of higher than 80%, on segments of speech of between 2s and 10s in length. AFRIKAANS : Suid-Afrika het elf amptelike tale waarvan tien as hulpbron-skaars beskou word. Vir die tien tale kan selfs die basiese hulpbronne wat benodig word om spraak tegnologie stelsels te ontwikkel moeilik wees om te bekom. Die proses om ‘n Gesproke Taal Identifisering stelsel vir hulpbron-skaars omgewings te ontwikkel, word in hierdie tesis ondersoek. ‘n Parallelle Foneem Herkenning gevolg deur Taal Modellering argitektuur word ingespan om drie spesifieke moontlikhede word ondersoek: (1) Onvolledige Hulpbronne, byvoorbeeld vermiste transkripsies en uitspraak woordeboeke; (2) Teenstrydige Hulpbronne, byvoorbeeld die gebruik van spraak data-versamelings wat teenstrydig is in terme van kanaal kenmerke; en (3) Hulpbronne van swak kwaliteit, byvoorbeeld foutief geklasifiseerde data en klank opnames wat swak getranskribeer is. Elke situasie word geanaliseer, tegnieke om die negatiewe effekte van min of swak hulpbronne te verminder word ontwikkel, en die bruikbaarheid van hierdie tegnieke word deur middel van eksperimente bepaal. Tegnieke wat ontwikkel word sluit die ontwikkeling van ortografiese ontleders, die outomatiese ontwikkeling van nuwe transkripsies, die filtrering van swak kwaliteit klank-data, klank-verdeling en kanaal normalisering tegnieke, en menslike verifikasie van verkeerd geklassifiseerde uitsprake in. Die kennis wat deur hierdie navorsing bekom word, word gebruik om die eerste Gesproke Taal Identifisering stelsel wat tussen al die tale van Suid-Afrika kan onderskei, te ontwikkel. Hierdie stelsel vaar relatief goed, en kan die elf tale met ‘n akkuraatheid van meer as 67% identifiseer. Indien daar op die ses taal families gefokus word, verbeter die persentasie tot meer as 80% vir segmente wat tussen 2 en 10 sekondes lank. Copyright / Dissertation (MEng)--University of Pretoria, 2010. / Electrical, Electronic and Computer Engineering / unrestricted

Ce travail expose plusieurs études dans les domaines de la reconnaissance de la parole et compréhension du langage parlé. La compréhension sémantique du langage parlé est un sous-domaine important de l'intelligence artificielle. Le traitement de la parole intéresse depuis longtemps les chercheurs, puisque la parole est une des charactéristiques qui definit l'être humain. Avec le développement du réseau neuronal artificiel, le domaine a connu une évolution rapide à la fois en terme de précision et de perception humaine. Une autre étape importante a été franchie avec le développement d'approches bout en bout. De telles approches permettent une coadaptation de toutes les parties du modèle, ce qui augmente ainsi les performances, et ce qui simplifie la procédure d'entrainement. Les modèles de bout en bout sont devenus réalisables avec la quantité croissante de données disponibles, de ressources informatiques et, surtout, avec de nombreux développements architecturaux innovateurs. Néanmoins, les approches traditionnelles (qui ne sont pas bout en bout) sont toujours pertinentes pour le traitement de la parole en raison des données difficiles dans les environnements bruyants, de la parole avec un accent et de la grande variété de dialectes. Dans le premier travail, nous explorons la reconnaissance de la parole hybride dans des environnements bruyants. Nous proposons de traiter la reconnaissance de la parole, qui fonctionne dans un nouvel environnement composé de différents bruits inconnus, comme une tâche d'adaptation de domaine. Pour cela, nous utilisons la nouvelle technique à l'époque de l'adaptation du domaine antagoniste. En résumé, ces travaux antérieurs proposaient de former des caractéristiques de manière à ce qu'elles soient distinctives pour la tâche principale, mais non-distinctive pour la tâche secondaire. Cette tâche secondaire est conçue pour être la tâche de reconnaissance de domaine. Ainsi, les fonctionnalités entraînées sont invariantes vis-à-vis du domaine considéré. Dans notre travail, nous adoptons cette technique et la modifions pour la tâche de reconnaissance de la parole dans un environnement bruyant. Dans le second travail, nous développons une méthode générale pour la régularisation des réseaux génératif récurrents. Il est connu que les réseaux récurrents ont souvent des difficultés à rester sur le même chemin, lors de la production de sorties longues. Bien qu'il soit possible d'utiliser des réseaux bidirectionnels pour une meilleure traitement de séquences pour l'apprentissage des charactéristiques, qui n'est pas applicable au cas génératif. Nous avons développé un moyen d'améliorer la cohérence de la production de longues séquences avec des réseaux récurrents. Nous proposons un moyen de construire un modèle similaire à un réseau bidirectionnel. L'idée centrale est d'utiliser une perte L2 entre les réseaux récurrents génératifs vers l'avant et vers l'arrière. Nous fournissons une évaluation expérimentale sur une multitude de tâches et d'ensembles de données, y compris la reconnaissance vocale, le sous-titrage d'images et la modélisation du langage. Dans le troisième article, nous étudions la possibilité de développer un identificateur d'intention de bout en bout pour la compréhension du langage parlé. La compréhension sémantique du langage parlé est une étape importante vers le développement d'une intelligence artificielle de type humain. Nous avons vu que les approches de bout en bout montrent des performances élevées sur les tâches, y compris la traduction automatique et la reconnaissance de la parole. Nous nous inspirons des travaux antérieurs pour développer un système de bout en bout pour la reconnaissance de l'intention. / This work presents several studies in the areas of speech recognition and understanding. The semantic speech understanding is an important sub-domain of the broader field of artificial intelligence. Speech processing has had interest from the researchers for long time because language is one of the defining characteristics of a human being. With the development of neural networks, the domain has seen rapid progress both in terms of accuracy and human perception. Another important milestone was achieved with the development of end-to-end approaches. Such approaches allow co-adaptation of all the parts of the model thus increasing the performance, as well as simplifying the training procedure. End-to-end models became feasible with the increasing amount of available data, computational resources, and most importantly with many novel architectural developments. Nevertheless, traditional, non end-to-end, approaches are still relevant for speech processing due to challenging data in noisy environments, accented speech, and high variety of dialects. In the first work, we explore the hybrid speech recognition in noisy environments. We propose to treat the recognition in the unseen noise condition as the domain adaptation task. For this, we use the novel at the time technique of the adversarial domain adaptation. In the nutshell, this prior work proposed to train features in such a way that they are discriminative for the primary task, but non-discriminative for the secondary task. This secondary task is constructed to be the domain recognition task. Thus, the features trained are invariant towards the domain at hand. In our work, we adopt this technique and modify it for the task of noisy speech recognition. In the second work, we develop a general method for regularizing the generative recurrent networks. It is known that the recurrent networks frequently have difficulties staying on same track when generating long outputs. While it is possible to use bi-directional networks for better sequence aggregation for feature learning, it is not applicable for the generative case. We developed a way improve the consistency of generating long sequences with recurrent networks. We propose a way to construct a model similar to bi-directional network. The key insight is to use a soft L2 loss between the forward and the backward generative recurrent networks. We provide experimental evaluation on a multitude of tasks and datasets, including speech recognition, image captioning, and language modeling. In the third paper, we investigate the possibility of developing an end-to-end intent recognizer for spoken language understanding. The semantic spoken language understanding is an important step towards developing a human-like artificial intelligence. We have seen that the end-to-end approaches show high performance on the tasks including machine translation and speech recognition. We draw the inspiration from the prior works to develop an end-to-end system for intent recognition.

Postkolonial teori har skiftat intresset från västerländska diskurser till frågor som ideologi, ojämlika maktförhållanden och etik. I samband med översättningsvetenskapens kulturella vändning på 1980-talet, började översättningsvetare ifrågasätta översättningsstrategier som antingen assimilerar (domesticering) eller stereotypiserar (exotisering) källkulturen. Newmark (1981) föreslår en semantisk, källtextorienterad översättningsprincip och menar att så länge den åstadkommer en likvärdig effekt, är en ordagrann översättning inte bara den föredragna, utan den enda godtagbara översättningsmetoden. Denna uppsats är en kommentar till min egen översättning av de första 17 kapitlen i romanen Welcome to Lagos, skriven av den nigerianska författaren Chibundu Onuzo. Källtexten har översatts med hjälp av en semantisk översättningsstrategi. Kommentaren fokuserar på tre aspekter som krävde särskild uppmärksamhet under översättningsarbetet, eftersom de utgör betydande utmaningar för semantiska överföringssätt: kulturspecifika begrepp, stilfigurer och talspråksmarkörer. I kommentaren framförs att den semantiska översättningsstrategin fungerade bra på den övergripande textnivån; även om specifika översättningsproblem ibland fick angripas med ett mer kommunikativt förhållningssätt för att åstadkomma en idiomatisk måltext med likvärdig effekt i målkulturen. / Postcolonial Studies shifted the interest from Western discourses to issues of ideology, power inequality, and ethics. As a consequence of the cultural turn in translation studies in the 1980s, scholars started questioning translation strategies that either assimilate (domestication) or stereotype (exoticization) the source culture. Proposing a semantic, source-text oriented translation principle, Newmark (1981) argues that as long as an equivalent effect can be achieved, literal translation is not just the preferred, but the only acceptable procedure. This paper comments on my own translation of the first 17 chapters of the novel Welcome to Lagos, written by Nigerian writer Chibundu Onuzo. The source text was translated using a semantic translation strategy. The commentary focuses on three key aspects that demanded particular attention during the translation process, due to the fact that they present significant challenges to semantic transfer methods: culture-specific items, stylistic devices, and spoken language markers. As the commentary suggests, the semantic translation strategy worked well on the global text level; occasionally, however, specific translation problems had to be dealt with using a more communicative approach in order to produce an idiomatic target text with an equivalent effect in the target culture.

In an increasingly interconnected world where English has become the lingua franca of business, culture, entertainment, and academia, learners of English as a second language (L2) have been steadily growing. This has contributed to an increasing demand for automatic spoken language assessment systems for formal settings and practice situations in Computer-Assisted Language Learning. One common misunderstanding about automated assessment is the assumption that machines should replicate the human process of assessment. Instead, computers are programmed to identify, extract, and quantify features in learners' productions, which are subsequently combined and weighted in a multidimensional space to predict a proficiency level or grade. In this regard, transferring human assessment knowledge and skills into an automatic system is a challenging task since this operation should take into account the complexity and the specificities of the proficiency construct. This PhD thesis presents research conducted on methods and techniques for the automatic assessment and feedback of L2 spoken English, mainly focusing on the application of deep learning approaches. In addition to overall proficiency grades, the main forms of feedback explored in this thesis are feedback on grammatical accuracy and assessment related to particular aspects of proficiency (e.g., grammar, pronunciation, rhythm, fluency, etc.). The first study explores the use of written data and the impact of features extracted through grammatical error detection on proficiency assessment, while the second illustrates a pipeline which starts from disfluency detection and removal, passes through grammatical error correction, and ends with proficiency assessment. Grammar, as well as rhythm, pronunciation, and lexical and semantic aspects, is also considered in the third study, which investigates whether it is possible to use systems targeting specific facets of proficiency analytically when only holistic scores are available. Finally, in the last two studies, we investigate the use of self-supervised learning speech representations for both holistic and analytic proficiency assessment. While aiming at enhancing the performance of state-of-the-art automatic systems, the present work pays particular attention to the validity and interpretability of assessment both holistically and analytically and intends to pave the way to a more profound and insightful knowledge and understanding of automatic systems for speaking assessment and feedback.