No description available.

No description available.

No description available.

No description available.

No description available.

Text-to-speech voices have come a long way in terms of their naturalness, and they are getting closer to human-sounding than ever. However, among the problems that still persist, the pronunciation of foreign words is still one of them. The experiments conducted in this thesis focus on using grapheme-to-phoneme (G2P) models to tackle the just-mentioned issue and, more specifically, to adjust the erroneous pronunciation of English words to an Italian English accent in Italian-speaking voices. We curated a dataset of words collected during recording sessions with an Italian voice actor reading general conversational sentences. We then manually transcribed their pronunciation in Italian English. In the second stage, we augmented the dataset by collecting the most common surnames in Great Britain and the United States, phonetically transcribed them with a rule-based phoneme mapping algorithm previously deployed by the company, and then manually adjusted the pronunciations to Italian English. Thirdly, by using the massively multilingual ByT5 model, a Transformer G2P model pre-trained on 100 languages, as well as its tokenizer-dependent versions T5_base and T5_small, and an LSTM with attention based on OpenNMT, we performed 10-fold cross-validation with the curated dataset. The results show that augmenting the data benefitted every model. In terms of PER, WER and accuracy, the transformer-based ByT5_small strongly outperformed its T5_small and T5_base counterparts even with a third or two-thirds of the training data. The second best performing model, the LSTM with attention one built with the OpenNMT framework, outperformed as well the T5 models, showed the second-best accuracy of our experiments and was the 'lightest' in terms of trainable parameters (2M) in comparison to ByT5 (299M) and the T5 ones (60 and 200M).

The literature on the influence of L2 orthography on abilingual’s pronunciation allows one to make predictions on what features of L2 pronunciation are likely to be impacted by it. This paper analyses the speech of 10 L1 Swedish L2 English late bilinguals performing one reading task and two tasks involving free speech in English. The speakers were predicted to realize /z/ as [s] when <s> represented /z/. All instances of /z/ spelled <s> were retrieved and transcribed in PRAAT. The number of substitutions for each speaker was calculated as a proportion of the total number of instances separately for reading and spontaneous speech. A 2-sample two-tailed z-test for proportions (p<0.05) was used to assess whether the two proportions were different. No difference was found in all speakers. This suggests that the degree of orthographic influence is fixed regardless of the presence of immediate orthographic input. Previous studies’ findings have contradicted these results, and this is discussed as a result of this study’s data containing connected speech as opposed to isolated words and phrases.

This thesis analyzes the phonological and morphological nativisation of English loans in the Tonga language. The contact situation between English and Tonga, in Zimbabwe, facilitates transference of lexical items between the two languages. From having been one of the most widely used languages of the world, English has developed into the most influential donor of words to other languages such as Tonga. The infiltration of English words into the Tonga lexical inventory led to the adoption and subsequent nativisation of English words by the native Tonga speakers. The main deposit of English words into Tonga is the direct interaction between English and Tonga speakers. However, it is sometimes via other languages like Shona, Ndebele, Venda and Shangani. In the 21st century, English’s contribution to the vocabulary of Tonga became more widely spread, now covering a large proportion of the Tonga language’s lexical inventory. The fact that English is the medium of instruction, in Zimbabwe, language of technology, education, media, new administration, health, music, new religion and economic transactions means that it is regarded as the high variety language with coercive loaning powers. Words from English are then adopted and nativised in the Tonga language, since Tonga asserts itself an independent language that can handle loans on its own. The main focus of this study therefore, is to try and account for the phonological and morphological behavior and changes that take place in English words that enter into Tonga. Analyzing phonological processes that are employed during nativisation of loan words entails analyzing how Tonga speakers handle aspects of English language such as diphthongs, triphthongs, cluster consonants, CVC syllable structure and sounds in repairing unacceptable sequences in Tonga. The research also accounts for the handling of morphological differences between the two languages. This entails looking at how competence and ordered-rule framework are harmonized by Tonga speakers in repairing conflicting features at morphological level. Since the two languages have different morphological patterns, the research analyzes the repairing strategies to handle singular and plural noun prefixes, tenses and particles, which are morphological components of words. The researcher appreciates that the native Tonga speakers have robust intuitions on the proper way to nativise words. / African Languages / D.Litt. et Phil. (African Languages)

Lam Wai Yung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (leaves 126-133). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Table of contents --- p.v / Chapter Chapter 1 --- INTRODUCTION --- p.1 / Chapter 1 .1 --- Background --- p.1 / Chapter 1.2 --- Purposes of The Study --- p.7 / Chapter Chapter 2 --- LITERATURE REVIEW --- p.8 / Chapter 2.1 --- Phonological Awareness and Reading --- p.8 / Chapter 2.1.1 --- Word recognition and reading --- p.8 / Chapter 2.1.2 --- Relationship between phonological awareness and reading performance --- p.9 / Chapter 2.2 --- Two Views About Levels of Phonological Awareness --- p.12 / Chapter 2.2.1 --- Linear view of levels of phonological awareness --- p.13 / Chapter 2.2.2 --- Hierarchical view of levels of phonological awareness --- p.14 / Chapter 2.3 --- Measurement of Phonological Awareness --- p.16 / Chapter 2.3.1 --- Various task types --- p.17 / Chapter 2.3.2 --- Various levels of difficulty of phonological tasks --- p.22 / Chapter 2.3.3 --- Task analysis studies --- p.23 / Chapter 2.4 --- Development of Levels of Phonological Awareness --- p.26 / Chapter 2.4.1 --- Developmental sequence of phonological awareness --- p.26 / Chapter 2.4.2 --- Underlying reasons for the sequential development --- p.29 / Chapter 2.5 --- Levels of Phonological Awareness and Reading --- p.30 / Chapter 2.5.1 --- Syllable awareness and English reading --- p.31 / Chapter 2.5.2 --- Phoneme awareness and English reading --- p.32 / Chapter 2.5.3 --- Onset-rime awareness and English reading --- p.34 / Chapter 2.6 --- Chinese Reading English as A Second Language --- p.39 / Chapter 2.6.1 --- Orthographic and phonological differences between Chinese and English --- p.39 / Chapter 2.6.2 --- Influence of learning Chinese on phonological awareness - --- p.41 / Chapter 2.6.3 --- Phonological awareness development in Chinese children - --- p.44 / Chapter 2.6.4 --- Effects of first language learning on reading English --- p.49 / Chapter Chapter 3 --- METHOD --- p.54 / Chapter 3.1 --- Subjects --- p.54 / Chapter 3.2 --- Tasks --- p.55 / Chapter 3.2.1 --- Syllable level tasks --- p.57 / Chapter 3.2.1.1 --- Syllable detection --- p.57 / Chapter 3.2.1.2 --- Syllable deletion --- p.58 / Chapter 3.2.2 --- Onset-rime level tasks --- p.58 / Chapter 3.2.2.1 --- Onset-rime detection --- p.58 / Chapter 3.2.2.2 --- Onset-rime deletion - --- p.59 / Chapter 3.2.3 --- Phoneme level tasks --- p.60 / Chapter 3.2.3.1 --- Phoneme detection --- p.60 / Chapter 3.2.3.2 --- Phoneme deletion - --- p.61 / Chapter 3.2.4 --- Word recognition test --- p.62 / Chapter 3.2.5 --- The Raven's Standard Progressive Matrices --- p.62 / Chapter 3.3 --- Procedure --- p.63 / Chapter Chapter 4 --- RESULTS --- p.64 / Chapter 4.1 --- Descriptive Data --- p.64 / Chapter 4.2 --- Patterns of Phonological Awareness --- p.66 / Chapter 4.2.1 --- The performance of three graders in the tasks of three levels of phonological awareness --- p.66 / Chapter 4.2.2 --- Patterns of phonological development --- p.76 / Chapter 4.3 --- Measuring Phonological Awareness: Detection Vs. Deletion --- p.82 / Chapter 4.4 --- Predicting English Word Reading Performance --- p.92 / Chapter 4.5 --- Summary of Results --- p.95 / Chapter Chapter 5 --- DISCUSSION --- p.98 / Chapter 5.1 --- Patterns of Phonological Awareness --- p.98 / Chapter 5.1.1 --- Progressive development in phonological awareness over grade --- p.98 / Chapter 5.1.2 --- The characteristics of phonological development in Hong Kong children --- p.103 / Chapter 5.1.3 --- Developmental progression from large via medium to small units --- p.107 / Chapter 5.2 --- Measurement of Phonological Awareness --- p.110 / Chapter 5.3 --- Phonological Awareness and Word Reading --- p.114 / Chapter Chapter 6 --- CONCLUSIONS AND RECOMMENDATIONS --- p.120 / Chapter 6.1 --- Major Findings --- p.120 / Chapter 6.1.1 --- Patterns of phonological awareness --- p.120 / Chapter 6.1.2 --- Measurement of phonological awareness --- p.120 / Chapter 6.1.3 --- Phonological awareness in predicting English reading --- p.121 / Chapter 6.2 --- Limitations --- p.122 / Chapter 6.3 --- Implications and Recommendations --- p.123 / Chapter 5.6.1 --- Pedagogical implications --- p.123 / Chapter 5.6.2 --- Future research recommendations --- p.124 / References --- p.126 / Appendixes --- p.134

本文主要探讨苏州话的连续变调，研究范围限定为音译词和复合词这两类多音节词。通过对苏州话语料的详细考察，本文总结出苏州话多音节词连读变调存在三层变调行为：（i）首音节尽量保留原调;（ii）第二个音节的变调与首音节的单字调有关;（iii）其余音节负载一个低平调. / 本文针对舒声调为首音节的多音节词的变调规律，回答了如下问题：（i）苏州话连续变调为何存在三层变调行为，而非两层或者四层？（ii）首音节保持原调的变调行为及它较高的负载声调的能力是由什么决定的？（iii）其余音节所负载的变调是如何决定的？ / 经过分析，我们将苏州话的变调域确认为韵律词，其内部结构如下：每个韵律词的头两个音节构成一个左重双拍音步，剩余的音节不构成音步因而直接被韵律词所管辖。由此苏州话的三层变调行为可以通过这三类音节在韵律词中的不同地位来解释，即：（i）首音节是左重双拍步中的强音节; （ii）第二个音节是左重双拍步中的弱音节; （iii）其余音节因不属于音步而被韵律词直接管辖。 / 基于对苏州话韵律结构的分析，本文以优选论为框架从理论上统一解释苏州话的连续变调。首先忠实性制约条件“禁止删除强音节单字调和“禁止强音节单字调变化 解释了首音节尽量保留原调的变调行为。其次，“禁止复杂仄调“，“禁止升调“和“禁止弱音节负载仄调这些简约性制约条件解释了现今苏州话连读变调模式中不允许出现复杂仄调，升调和非首音节仄调的现象。再次，“尾音节连接低调解释了连读变调后韵律词尾音节所负载的低平调。 / 本文通过对音译词和复合词的分类考察，补充了前人对苏州话连读变调现象的描述。而且在以往理论研究的基础上，本文论证了苏州话的连读变调是由它的韵律结构所决定的，由此补足了苏州话变调现象的理论解释。最后，本文对苏州话的分析也进一步印证了韵律结构的层级关系和韵律单位是人类语言所共有的，但韵律单位的构建在不同语言中则各有不同。 / This thesis presents an optimality-theoretic account of Suzhou tone sandhi from a prosodic perspective. By investigating transliterations and compounds, we find a three-way distinction of tone sandhi behavior within a tone sandhi domain: (i) the initial syllable has the ability to retain its citation tone; (ii) the second syllable carries a sandhi level tone related to the citation tone in the initial syllable; and (iii) the remaining syllables carry a low level tone. / Focusing on the tone sandhi patterns with initial long tones, we answer the following research questions: (i) what determines the existence of a three-way distinction of tone sandhi behavior, rather than a two-way or four-way one? (ii) what determines the tone stability and the greater tone-bearing ability of the initial syllable? and (iii) how are the sandhi tones in the non-initial syllables determined? / The tone sandhi domain in Suzhou Chinese is identified as the prosodic word, which contains a single left-headed binary foot and unfooted syllables. The three-way distinction is then captured by the three prosodic states in a prosodic word: (i) the strong syllable in the left-headed binary foot within the prosodic word, (ii) the weak syllable in this foot, and (iii) the syllables which are unfooted and immediately dominated by the prosodic word. / Based on the analysis of Suzhou prosodic structure, our account of Suzhou tone sandhi is formulated within the framework of Optimality Theory. First, ranking the positional faithfulness constraints with reference to strong syllables higher than the context-free faithfulness constraints guarantees the tone stability of the initial syllables. Second, the undominated markedness constraints which require no complex contour tone and no rising tone trigger contour tone reduction and tone redistribution. Another undominated markedness constraint which requires no contour tone in weak syllables ensures no contour tones in a weak syllable, i.e. a non-initial syllable in Suzhou Chinese. Third, the final low level tone in the sandhi patterns is determined by the markedness constraint which requires low tone assignment in the rightmost syllable of a prosodic word. / This thesis complements the descriptive studies on Suzhou tone sandhi, by adding the data of transliterations and investigating compounds according to the morphological relations. It also confirms the idea in previous theoretical studies on Chinese languages that prosodic structure governs tone sandhi, including the tone stability in strong syllables and the greater tone-bearing ability of strong syllables. This thesis further corroborates the idea in Prosodic Phonology that the prosodic hierarchy and its constituent are universal, though the precise shape of the constituent are language-specific. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Shi, Xinyuan. / "November 2012." / Thesis (M.Phil.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references. / Abstracts also in Chinese. / ABSTRACT --- p.I / ACKNOWLEDGEMENTS --- p.I / Chapter CHAPTER 1 --- INTRODUCTION --- p.1 / Chapter 1.1 --- OVERVIEW --- p.1 / Chapter 1.2 --- PRELIMINARY: PHONETIC INVENTORIES OF SUZHOU CHINESE --- p.2 / Chapter 1.3 --- THEORETICAL FRAMEWORK: OPTIMALITY THEORY --- p.6 / Chapter 1.3.1 --- Basic principles --- p.6 / Chapter 1.3.2 --- Conflicts between markedness and faithfulness --- p.9 / Chapter 1.3.3 --- Alignment constraints --- p.9 / Chapter 1.4 --- ORGANIZATION --- p.10 / Chapter CHAPTER 2 --- INVESTIGATION OF TONE SANDHI PATTERNS IN SUZHOU CHINESE --- p.11 / Chapter 2.1 --- DESCRIPTIVE STUDIES ON TONAL PHONOLOGY OF SUZHOU CHINESE --- p.11 / Chapter 2.1.1 --- Citation tones --- p.11 / Chapter 2.1.2 --- Tone sandhi --- p.15 / Chapter 2.1.3 --- Interim summary --- p.21 / Chapter 2.2 --- TONE SANDHI PATTERNS IN SUZHOU CHINESE: THE CURRENT INVESTIGATION . --- p.22 / Chapter 2.2.1 --- Tone sandhi patterns in transliterations --- p.22 / Chapter 2.2.1.1 --- Tone sandhi patterns beginning with long tones --- p.23 / Chapter 2.2.1.2 --- Tone sandhi patterns beginning with short tones --- p.28 / Chapter 2.2.2 --- Tone sandhi patterns in compounds --- p.32 / Chapter 2.2.2.1 --- Tone sandhi patterns in compounds I: modification --- p.34 / Chapter 2.2.2.2 --- Tone sandhi patterns in compounds II: coordination --- p.38 / Chapter 2.2.2.3 --- Tone sandhi patterns in compounds III: other morphological relations --- p.40 / Chapter 2.2.2.4 --- Intra-variations in disyllabic compounds: [HL.L]~[HL.H] --- p.42 / Chapter 2.2.2.5 --- Inter-speaker variations in compounds: [LL.H]~[LH.H] --- p.44 / Chapter 2.2.2.6 --- Tone sandhi in a group of frequently used compounds --- p.46 / Chapter 2.2.3 --- Tone sandhi in fast tempo --- p.50 / Chapter 2.3 --- SUMMARY --- p.53 / Chapter CHAPTER 3 --- PREVIOUS STUDIES OF TONE SANDHI IN CHINESE LANGUAGES --- p.57 / Chapter 3.1 --- STUDIES OF THE REPRESENTATION OF TONE WITHIN THE SPE FRAMEWORK --- p.57 / Chapter 3.1.1 --- Wang’s (1967) ground-breaking proposal of tone features --- p.58 / Chapter 3.1.2 --- Woo’s (1969) proposal of decomposing contour tones --- p.59 / Chapter 3.1.3 --- Yip’s (1980) proposal of two pitch heights in each register --- p.61 / Chapter 3.1.4 --- Interim summary --- p.63 / Chapter 3.2 --- STUDIES OF TONE SANDHI IN NORTHERN WU WITHIN THE METRICAL-AUTOSEGMENTAL FRAMEWORK --- p.64 / Chapter 3.2.1 --- Duanmu’s (1990) moraic analysis of tone re-association in New Shanghai --- p.65 / Chapter 3.2.2 --- Ao’s (1993) identification of tone sandhi domain in Nantong Chinese --- p.66 / Chapter 3.2.3 --- Interim summary --- p.69 / Chapter 3.3 --- STUDIES OF TONAL DISTRIBUTION WITHIN OT FRAMEWORK --- p.71 / Chapter 3.3.1 --- Jiang-King’s (1996) tone-syllable weight correlation in Northern Min --- p.71 / Chapter 3.3.2 --- Zhang’s (2001) tone-duration correlation in typological survey --- p.73 / Chapter 3.3.3 --- Interim summary --- p.74 / Chapter 3.4 --- SUMMARY --- p.76 / Chapter CHAPTER 4 --- AN OPTIMALITY THEORETICAL ACCOUNT OF TONE SANDHI IN SUZHOU CHINESE --- p.77 / Chapter 4.1 --- IDENTIFYING THE TONE SANDHI DOMAIN --- p.78 / Chapter 4.1.1 --- Identifying prosodic word as the tone sandhi domain --- p.78 / Chapter 4.1.1.1 --- Brief introduction of the prosodic hierarchy --- p.78 / Chapter 4.1.1.2 --- Prosodic categories relevant to Suzhou tone sandhi --- p.80 / Chapter 4.1.1.3 --- Construction of prosodic word in Suzhou Chinese --- p.82 / Chapter 4.1.2 --- Identifying morpho-syntactic unit as the tone sandhi domain --- p.86 / Chapter 4.1.2.1 --- Inadequacy of identifying morphological unit as the tone sandhi domain --- p.89 / Chapter 4.1.2.2 --- The relation between tone sandhi domains and morphological units --- p.91 / Chapter 4.1.2.3 --- Mismatch between tone sandhi domain and syntactic unit --- p.93 / Chapter 4.1.3 --- Interim summary --- p.95 / Chapter 4.2 --- CONSTRAINTS ON SUZHOU TONE SANDHI PATTERNS --- p.96 / Chapter 4.2.1 --- Constraints on the structure of the prosodic word --- p.96 / Chapter 4.2.2 --- Constraints on tone retention --- p.99 / Chapter 4.2.3 --- Constraints on contour tone prohibition --- p.101 / Chapter 4.2.4 --- Constraint on tonal assignment --- p.102 / Chapter 4.3 --- PREDICTING TONE SANDHI PATTERNS BY CONSTRAINT RANKING --- p.104 / Chapter 4.3.1 --- Predicting tone sandhi patterns in transliterations --- p.104 / Chapter 4.3.2 --- Predicting tone sandhi patterns in compounds --- p.110 / Chapter 4.3.2.1 --- Intra-speaker variations in compounds: [HL.L]~[HL.H] --- p.110 / Chapter 4.3.2.3 --- Inter-speaker variations in compounds: [LL.H]~[LH.L] --- p.113 / Chapter 4.3.2.4 --- Tone sandhi in fast speech tempo --- p.117 / Chapter 4.4 --- SUMMARY --- p.119 / Chapter CHAPTER 5 --- CONCLUSION --- p.120 / LIST OF REFERENCES --- p.125