This study investigated the language use of deaf adult bilinguals in conversation with each other in workplace settings, and with their deaf and hearing children in home settings. The aim was to gain insight into the Auslan-English language contact outcomes that might be found in these settings, and what factors influenced these outcomes. The results indicated that the most unique use of language by deaf bilinguals was that of simultaneous use of both spoken English and Auslan, and it was this simultaneous use which facilitated the two examples of code-switching (defined as a complete change of language from Auslan to spoken English) that was found in the data. The other two contact outcomes of significance were frequent transference of English into Auslan, and the equally frequent use of fingerspelling, which has a pivotal role in filling the gap in Auslan, a language with no orthographic form. The study also revealed that Auslan (a signed language) was the language in which many issues of identity were expressed by deaf bilinguals, regardless of whether the individual was a first or second language learner of Auslan. The results confirmed that these language and identity factors did influence the language contact outcomes.

Thesis (MA)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: The use of interpersonal verbal routines such as greetings is a universal phenomenon of human languages. All human speech communities have such formulas, although their character and the incidence of their use may vary enormously from one society to another. For several decades, greetings have been a recurrent object of inquiry for linguists and other human communication. Greetings are part of phatic communion, whereby people create ties of union and avoid silence, which is always alarming and dangerous. Communion among humans will often be marked in speech “phatically”. There is widespread evidence that greetings are an important part of the communicative competence necessary for being a member of any speech community. Greetings regularize patterns among members. Greeting has been often treated as if it was spontaneous emotional reaction to the coming together of people carrying overtly its own social message. Greeting expressions constitute an important part of the polite language. By greeting the speaker, indicates his attitude towards the addressee or starts a conversation with him. Greetings are often patterned expressions, which may vary among different nations. Most greetings perform primarily a phatic communion function; some greetings are used to convey information. Some culture does not operate non-verbal demonstration of respect of difference like bowing, or prostrating and kneeling, it makes up for this by insistence on the proper execution of verbal greetings, for example: Igbo culture does not operate nonverbal, whereas Japanese, Joruba and even Vendas they do practice these non-verbal demonstrations. Cultural performances are influenced by social variables such as the ages, sex and status of the interactants. / AFRIKAANSE OPSOMMING: Die gebruik van interpersoonlike mondelingse roetines, soos die handeling van groet, is ‘n universiele fenomeen van menslike taal. Alle menslike gemeenskappe het formules, alhoewel hulle karakter en die voorkoms van hulle gebruik, mag verskil van een gemeenskap tot ‘n ander. Vir dekades, was groet ‘n herhaalde onderwerp van ondersoek van taalkenners saam met ander aspekte van persoonlike kommunikasie. Die handeling van groet is ‘n deel van fatiese kommunikasie, waarby mense bande skep, en stilte vermy wat angswekkend kan wees. Kommunikasie tussen mense sal altyd na verwys word as faties in taalverskynsels. Daar is wydverspreide bewyse dat die handeling van groet ‘n belangrike deel van kommunikatiewe kompetensie is, wat noodsaaklik is vir 'n lid van enige gemeenskap geld. Groet reguleer voorbeelde van wedersydse verhoudings tussen groepelede. Groet is dikwels hanteer asof dit ‘n spontane emosionele reaksie by die saamkom van mense is wat hulle eie sosiale boodskap oordra. Die spraakhandeling van groet vorm ‘n belangrike deel van beleefdheidtaal. Deur te groet, bewys die persoon sy houding teenoor die ander persoon of begin om met die persoon 'n gesprek te voer. Die spraakhandeling van groet is dikwels voorbeelde van uitdrukkings wat verskil tussen verskillende taalgroepe. Die meeste groetvorme het ‘n primere fatiese gemeenskaps funksie, sommige begroetings word gebruik om informasie te verskaf. Sommige kulture maak nie gebruik van nie-verbale demonstrasies van respek of verskille soos neerbuiging of kniel, dit maak op vir die aandring op behoorlike gebruik van mondelinge begroeting, byvoorbeeld: Igbo kultuur maak nie gebruik van nie-verbale demonstrasies, waar Vendakultuur gebruik maak van hierdie nie-verbale demonstrasies. Kulturele belewenis van die groetvorm word beVnvloed deur sosiale veranderlikes soos ouderdom, geslag en status.

published_or_final_version / Chinese / Master / Master of Philosophy

published_or_final_version / abstract / Linguistics / Master / Master of Arts in Applied Linguistics

published_or_final_version / Applied English Studies / Master / Master of Arts in Applied Linguistics

In spoken French, the phonological processes of liaison and resyllabification can render word and syllable boundaries ambiguous. In the case of liaison, for example, the /n/ in the masculine indefinite article un [oẽ] is normally latent, but when followed by a vowel-initial word the /n/ surfaces and is resyllabified as the onset of that word. Thus, the phrases un air ‘a melody’ and un nerf ‘a nerve’ are produced with identical phonemic content and syllable boundaries [oẽ.nɛʁ]). Some research has suggested that speakers of French give listeners cues to word boundaries by varying the duration of consonants that surface in liaison environments relative to consonants produced word-initially. Production studies (e.g. Wauquier-Gravelines 1996; Spinelli et al. 2003) have demonstrated that liaison consonants (e.g. /n/ in un air) are significantly shorter than the same consonant in initial position (e.g. /n/ in un nerf). Studies on the perception of spoken French have suggested that listeners exploit these durational differences in the segmentation of running speech (e.g. Gaskell et al. 2002; Spinelli et al. 2003), though no study to date has tested this hypothesis directly. The current study employs a direct test of the exploitation of duration as a segmentation cue by manipulating this single acoustic factor while holding all other factors constant. Thirty-six native speakers of French and 54 adult learners of French as a second language (L2) were tested on both an AX discrimination task and a forced-choice identification task which employed stimuli in which the durations of pivotal consonants (e.g. /n/ in [oẽ.nɛʁ]) were instrumentally shortened and lengthened. The results suggest that duration alone can indeed modulate the lexical interpretation of sequences rendered sequences in spoken French. Shortened stimuli elicited a significantly larger proportion of vowel-initial (liaison) responses, while lengthened stimuli elicited a significantly larger proportion of consonant-initial responses, indicating that both native and (advanced) non-native speakers are indeed sensitive to this acoustic cue. These results add to a growing body of work demonstrating that listeners use extremely fined-grained acoustic detail to modulate lexical access (e.g. Salverda et al. 2003; Shatzman & McQueen 2006). In addition, the current results have manifest ramifications for study of the upper limits of L2 acquisition and the plasticity of the adult perceptual system in that they show evidence nativelike sensitivity to non-contrastive phonological variation. / text

在兒童習得談話溝通的發展過程中，兒童需要習得多種談話技巧才能全然地參與談話，而其中一項重要的談話技巧即為引語的使用。引語在言談分析中是指一種半固定的語句用以加速言談的流暢。然而至今沒有研究探討過兒童使用引語的情形，因此本研究旨在探討中文言談中學齡兒童使用引語的發展情形。本研究將十二名學齡兒童依年齡分為兩組，並將其自然談話內容錄音且加以轉錄，以資進一步之分析與研究。此研究結果發現隨著年紀的增長兒童愈常使用引語，且愈能行使不同的種類形式來表達。這些發展性的成長與兒童的心智、語言及社會發展有關；此外，本研究結果亦發現學齡兒童最常使用的引語種類。本研究期待能增加人們對於兒童引語的認識，更能提供國小教師指導或幫助學齡兒童在言談中使用引語，進而增進兒童言談順暢無誤。

The primary claim of this dissertation is that children and adults process language in the same manner, meaning that when children are acquiring their first language what they are truly doing is perfecting their language processing abilities. Language acquisition and processing both start from the same place. Both work to find patterns in the signal that will, eventually, be paired with meaning. This dissertation argues that differences in how children and adults accomplish these tasks are one of degree and not kind. To show this, three experiments tested how adults and children responded to a conflict between the lexical and prosodic parse of an utterance. The participants’ response to this conflict reveals information about where they are in the language acquisition process. In these experiments, prosody was used to disambiguate phrases that can be interpreted either as a list of two items (e.g., fruit, salad) or as a single compound item (e.g., fruit-salad). Prosody was also made to conflict with the lexical parse of an utterance. When the word cactus is said with List Prosody two non-words /kæk/ and /tʌs/ result. When the words nail and key are said with Compound Prosody, the non-word nailkey is created. By exploiting the overlap between the prosodic system and the lexical system, it is possible to evaluate how language is being processed. The results show that adults tend to parse utterances based on the lexical content, and ignore ambiguities created by a conflict between the prosodic and the lexical interpretation of the phrase. In contrast, children tend to respond based on the prosody, making increasing use of the lexical content as they mature. When the same items are tested with abstract shapes rather than representational images, adults make greater use of prosody. This suggests that visual input plays a role in spoken word processing. The dissertation also proposes a modified model of spoken word recognition that accounts for the difference seen between the adults and the children, and for the effect of visual content. This model integrates phonetic details, prosodic content, lexical knowledge, visual content, and pragmatic understanding during spoken word recognition.

We address the problem of dynamic user modelling for referring expression generation in spoken dialogue systems, i.e how a spoken dialogue system should choose referring expressions to refer to domain entities to users with different levels of domain expertise, whose domain knowledge is initially unknown to the system. We approach this problem using a statistical planning framework: Reinforcement Learning techniques in Markov Decision Processes (MDP). We present a new reinforcement learning framework to learn user modelling strategies for adaptive referring expression generation (REG) in resource scarce domains (i.e. where no large corpus exists for learning). As a part of the framework, we present novel user simulation models that are sensitive to the referring expressions used by the system and are able to simulate users with different levels of domain knowledge. Such models are shown to simulate real user behaviour more closely than baseline user simulation models. In contrast to previous approaches to user adaptive systems, we do not assume that the user’s domain knowledge is available to the system before the conversation starts. We show that using a small corpus of non-adaptive dialogues it is possible to learn an adaptive user modelling policy in resource scarce domains using our framework. We also show that the learned user modelling strategies performed better in terms of adaptation than hand-coded baselines policies on both simulated and real users. With real users, the learned policy produced around 20% increase in adaptation in comparison to the best performing hand-coded adaptive baseline. We also show that adaptation to user’s domain knowledge results in improving task success (99.47% for learned policy vs 84.7% for hand-coded baseline) and reducing dialogue time of the conversation (11% relative difference). This is because users found it easier to identify domain objects when the system used adaptive referring expressions during the conversations.

Spoken term detection (STD) is a fundamental task for multimedia information retrieval. A major challenge faced by an STD system is the serious performance reduction when detecting out-of-vocabulary (OOV) terms. The difficulties arise not only from the absence of pronunciations for such terms in the system dictionaries, but from intrinsic uncertainty in pronunciations, significant diversity in term properties and a high degree of weakness in acoustic and language modelling. To tackle the OOV issue, we first applied the joint-multigram model to predict pronunciations for OOV terms in a stochastic way. Based on this, we propose a stochastic pronunciation model that considers all possible pronunciations for OOV terms so that the high pronunciation uncertainty is compensated for. Furthermore, to deal with the diversity in term properties, we propose a termdependent discriminative decision strategy, which employs discriminative models to integrate multiple informative factors and confidence measures into a classification probability, which gives rise to minimum decision cost. In addition, to address the weakness in acoustic and language modelling, we propose a direct posterior confidence measure which replaces the generative models with a discriminative model, such as a multi-layer perceptron (MLP), to obtain a robust confidence for OOV term detection. With these novel techniques, the STD performance on OOV terms was improved substantially and significantly in our experiments set on meeting speech data.