In this thesis, I investigate a decades-old problem found in the stop system of Proto-Indo-European (PIE). More specifically, I will be investigating the paucity of */b/ in the forms reconstructed for the ancient, hypothetical language. As cross-linguistic evidence and phonological theory alone have fallen short of providing a satisfactory answer, herein will I employ modern empirical methods of linguistic investigation, namely laboratory phonology experiments and computational database analysis. Following Byrd 2015, I advocate for an examination of synchronic phenomena and behavior as a method for investigating diachronic change. In Chapter 1, I present an overview of the various proposed phonological systems of PIE and some of the explanations previously given for the enigmatic rarity of PIE */b/. Chapter 2 presents a detailed account of three lab phonology experiments I conducted in order to investigate perceptual confusability as a motivator of asymmetric merger within a system of stop consonants. Chapter 3 presents the preliminary form and findings of a computational database of reconstructed forms in PIE that I created and have named the Database of Etymological Reconstructions Beginnning in Proto-Indo-European (DERBiPIE). The final chapter, Chapter 4, offers a summary of the work presented herein and conclusions that may be drawn, offering suggestions for continued work on the topic and others like it.

This thesis investigates issues of intelligibility through the lens and focus of prosody when the Bostonian and Los Angeles-based accents are heard in casually occurring conversation by native and non-native speakers. Over the spring and summer of 2017, six native speakers and 11 non-native speakers of English were interviewed from having listened to two 2.5 minute audio sample clips of speakers who have these accents. Respondents were asked questions such as what was difficult or easy or whether they could summarize the recordings for me. Findings indicate that while the native speakers often had difficulty with vocabulary due to context, non-native English speakers frequently found the same recordings to sound continuous, blended or merged together when the Boston and Los Angeles audios were played to them. Native English speakers, by contrast, did not seem to face the same prosodic challenges of intelligibility as their non-native English-speaking counterparts when these two accents were heard in informal conversation. It has been found that L2 learners have a strong desire to learn English from their teachers through more naturally or informally occurring conversation. The argument is made that the teaching, practice and engagement of informal conversation is woefully inadequate for non-native speakers of English. Within this thesis the core subsets of the perceptions of prosody are analyzed between native and non-native speakers of English. The purpose of doing so is to pedagogically improve learning in EFL and ESL contexts.

This study investigated whether East Asian learners of English (n=8) studying in the US acquired more accurate intonation patterns (compared to native-speaker norms) after receiving five weeks of tutoring focusing on four basic intonation patterns (definite statements, wh-questions, yes/no questions, and tag questions) and using oral reading as the primary practice technique. The study also assessed the students' affective reaction to the teaching method through interviews. The study found that the learners significantly improved their intonational accuracy (based on the judgments of three native speakers who listened to single-sentence recordings [n=868] from questionnaires, exit interviews, and pre- and post-tests) and that they were generally amenable to the teaching technique.

This dissertation examines the relationship between phonetic sound and content in Latin poetry, with a focus on Books 1-3 of Horace's Odes. The central argument is that a relationship exists between sound and content in poetry, that this can be analysed and described more thoroughly and systematically than is usually the case, and that the appreciation of poetry can be enhanced by doing so. Part 1 presents a scheme for describing the sound-content relationship, and argues that this accurately reflects the perceptions of poetic audiences and is psychologically valid. The scheme begins with the concept of the 'sonance', defined as any set of sounds that renders a passage sonically noteworthy. Sonances that relate to content are classified either as 'harmonic sonances', which relate to content due to the properties of the relevant sounds, or 'repetitive sonances', which relate to content purely due to the repetition (including patterning or contrasting) of sounds, regardless of their properties. Harmonic sonances, it is argued, may relate to content through four 'harmonies', depending on whether acoustic or articulatory properties are involved and whether the relationship is one of similarity between property and content or a more distant 'metaphor'. Repetitive sonances may relate to content by several different means, or 'modes of repetition'. Part 2 presents a simple method of numerical analysis which may be applied to the text by computer to extract passages that are relatively likely to contain a sonance, and briefly discusses the process of assessing these results, identifying further sonances by more natural means, and relating each sonance to the relevant content. As an essential preliminary to such matters, Latin phonetics and phonology are also discussed in detail. Parts 3 and 4 present the results of applying these resources of assessment and description to the text, to demonstrate the type of poetic appreciation which may thus be gained. Part 3 consists of two catalogues of harmonic and repetitive sonances taken from the whole of Odes 1-3. Part 4 is a specific examination of two entire odes in much greater detail. / Thesis (Ph.D.)--Humanities, 2004.

<p>Learning a second language is different from learning our first one. A lot of rules from the first language, concerning e.g. grammar, intonation and phonology, are so firmly rooted within learners that they will transfer them to the new language regardless of whether they are correct or not. Studies show that the way we are tuned in to the sounds of our first language can make it difficult for us to perceive the phonemes of a new language correctly. In order to study the relationship between Swedish speakers’ faulty production of English vowels and their perception of them, ten subjects participated in a perception test to find out how well they could distinguish between minimal pairs containing phonemes that Swedes often have problems pronouncing correctly. They were also recorded while reading sentences containing the same minimal pairs. The results from the perception test were compared to graphs showing how consistent the subjects were in their pronunciation of these phonemes. The study shows that although some phonemes proved to be more difficult for the subjects to perceive a difference between, a faulty production of these sounds cannot be explained by misperception alone.</p>

<p>This study investigates the realisation of phonological quantity in the dialects of Modern Swedish, based on a corpus containing recordings from 86 locations in Sweden and the Swedishspeaking parts of Finland. The corpus was recorded as part of the national SweDia project.</p><p>The study is explorative in character. Quantity structures in Swedish dialects and their geographical distribution, as described in the dialectological literature, are compared to the results of a data-driven categorisation (cluster analysis). The results reveal an overall good correspondence of the data driven and the traditional categorisation, although with some deviations in the detail.</p><p>The study is divided into two parts. The first part lays the foundation for the data-driven categorisation, which is then described in the second part. First, the phonology and phonetics of quantity in Swedish are described in terms of durational distinctions and vocalic quality differences that typically accompany the durational differences. Preaspiration, which appears to be a normative feature in some dialects, is covered as well. An overview of the historical development of the Swedish quantity system is provided, with special emphasis on a phonological interpretation of quantity changes. Thereafter, dialectological evidence is combined with phonological and typological considerations to develop a categorisation of Swedish dialects.</p><p>The second part explains the methodology of cluster analysis and applies this method to vowel and consonant durations from one contrastive word pair, in order to obtain an alternative dialect categorisation. Analyses of vowel quality and preaspiration are performed in addition to the durational analyses. Hypotheses derived from the cluster analysis are then tested on one additional word pair recorded in 75 locations and on three additional word pairs recorded in four locations.</p><p>The general pattern emerging from the cluster analysis is a categorisation of the dialects into three main types, a Finland-Swedish, a Northern and a Southern type. This categorisation shows a good geographical agreement with the categorisation that is derived from the analysis of the dialectological literature. Therefore, the durational patterns of the three types are interpreted as reflections of three different phonological systems: 4-way systems with vocalic and consonantal quantity, 3-way systems with vocalic quantity and with consonantal quantity only after short vowels, and 2-way systems with complementary quantity. From the historical perspective, the 4-way system constitutes the most conservative and the 2-way system the most recently developed system.</p><p>Finally, it is argued that the historical development is one of the factors behind occasional mismatches between the data-driven and the dialectological categorisation. Data from one of the dialects, which has recently abandoned a 4-way system but has obviously retained the durational properties of the older system, is used as an example to illustrate this historical hypothesis.</p>

This study investigates the realisation of phonological quantity in the dialects of Modern Swedish, based on a corpus containing recordings from 86 locations in Sweden and the Swedishspeaking parts of Finland. The corpus was recorded as part of the national SweDia project. The study is explorative in character. Quantity structures in Swedish dialects and their geographical distribution, as described in the dialectological literature, are compared to the results of a data-driven categorisation (cluster analysis). The results reveal an overall good correspondence of the data driven and the traditional categorisation, although with some deviations in the detail. The study is divided into two parts. The first part lays the foundation for the data-driven categorisation, which is then described in the second part. First, the phonology and phonetics of quantity in Swedish are described in terms of durational distinctions and vocalic quality differences that typically accompany the durational differences. Preaspiration, which appears to be a normative feature in some dialects, is covered as well. An overview of the historical development of the Swedish quantity system is provided, with special emphasis on a phonological interpretation of quantity changes. Thereafter, dialectological evidence is combined with phonological and typological considerations to develop a categorisation of Swedish dialects. The second part explains the methodology of cluster analysis and applies this method to vowel and consonant durations from one contrastive word pair, in order to obtain an alternative dialect categorisation. Analyses of vowel quality and preaspiration are performed in addition to the durational analyses. Hypotheses derived from the cluster analysis are then tested on one additional word pair recorded in 75 locations and on three additional word pairs recorded in four locations. The general pattern emerging from the cluster analysis is a categorisation of the dialects into three main types, a Finland-Swedish, a Northern and a Southern type. This categorisation shows a good geographical agreement with the categorisation that is derived from the analysis of the dialectological literature. Therefore, the durational patterns of the three types are interpreted as reflections of three different phonological systems: 4-way systems with vocalic and consonantal quantity, 3-way systems with vocalic quantity and with consonantal quantity only after short vowels, and 2-way systems with complementary quantity. From the historical perspective, the 4-way system constitutes the most conservative and the 2-way system the most recently developed system. Finally, it is argued that the historical development is one of the factors behind occasional mismatches between the data-driven and the dialectological categorisation. Data from one of the dialects, which has recently abandoned a 4-way system but has obviously retained the durational properties of the older system, is used as an example to illustrate this historical hypothesis.

Speech communication encompasses diverse types of information, including phonetics, affective state, voice quality, and speaker identity. From a speech production standpoint, the acoustic speech signal can be mainly divided into glottal source and vocal tract components, which play distinct roles in rendering the various types of information it contains. Most deployed speech analysis systems, however, do not explicitly represent these two components as distinct entities, as their joint estimation from the acoustic speech signal becomes an ill-defined blind deconvolution problem. Nevertheless, because of the desire to understand glottal behavior and how it relates to perceived voice quality, there has been continued interest in explicitly estimating the glottal component of the speech signal. To this end, several inverse filtering (IF) algorithms have been proposed, but they are unreliable in practice because of the blind formulation of the separation problem. In an effort to develop a method that can bypass the challenging IF process, this thesis proposes a new glottal source information extraction method that relies on supervised machine learning to transform smoothed spectral representations of speech, which are already used in some of the most widely deployed and successful speech analysis applications, into a set of glottal source features. A transformation method based on Gaussian mixture regression (GMR) is presented and compared to current IF methods in terms of feature similarity, reliability, and speaker discrimination capability on a large speech corpus, and potential representations of the spectral envelope of speech are investigated for their ability represent glottal source variation in a predictable manner. The proposed system was found to produce glottal source features that reasonably matched their IF counterparts in many cases, while being less susceptible to spurious errors. The development of the proposed method entailed a study into the aspects of glottal source information that are already contained within the spectral features commonly used in speech analysis, yielding an objective assessment regarding the expected advantages of explicitly using glottal information extracted from the speech signal via currently available IF methods, versus the alternative of relying on the glottal source information that is implicitly contained in spectral envelope representations.

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

Chinese / Doctoral / Doctor of Philosophy