No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
The English pronunciation of Arabic speakers: A data-driven approach to segmental error identification
Abstract
The accurate identification of likely segmental pronunciation errors produced by nonnative speakers of English is a longstanding goal in pronunciation teaching. Most lists of pronunciation errors for speakers of a particular first language (L1) are based on the experience of expert linguists or teachers of English as a second language (ESL) and English as a foreign language (EFL). Such lists are useful, but they are also subject to blind spots for less noticeable errors while suggesting that other more noticeable errors are more important. This exploratory study tested whether using a database of read sentences would reveal recurrent errors that had been overlooked by expert opinions. We did a systematic error analysis of advanced L1 Arabic learners of English ( n = 4) using L2 Arctic, a publicly available collection of 1,132 phonetically-balanced English sentences read aloud by 24 speakers of six language backgrounds. To test whether the database was useful for pronunciation error identification, we analysed Arabic speakers’ sentence readings ( n = 599), which were annotated in Praat for pronunciation deviations from General American English. The findings give an empirically supported description of persistent pronunciation errors for Arabic learners of English. Although necessarily limited in scope, the study demonstrates how similar datasets can be used regardless of the L1 being investigated. The discussion of errors in pronunciation in terms of their functional loads (Brown, 1988) suggests which persistent errors are likely to be important for classroom attention, helping teachers focus their limited classroom time for optimal learning.
... Teachers can use multimodality. The synergistic effect of attitude enables students to understand the characteristics of English pronunciation from hearing, vision, and touch and improve English pronunciation. Figure 1 shows the multimodel [1][2][3][4][5][6][7][8][9][10]. ...
In recent years, with the increasing frequency of international exchanges, people have gradually realized that language is a tool of communication and communication, and language learning should attach importance to oral teaching. However, in traditional classrooms, one of the problems faced by oral teaching is the mismatch of the teacher-student ratio: a teacher has to deal with dozens of students, one-on-one oral teaching and pronunciation guidance is impossible, and it is also affected by the teachers and the environment constraints. Therefore, the research on how to efficiently automate pronunciation training is becoming more and more popular. Many phonemes in English have different facial visual features, especially vowels. Almost all of them can be distinguished by the roundness and tightness of the lips in appearance. In order to give full play to the role of lip features in oral pronunciation error detection, this paper proposes a multimodal feature fusion model based on lip angle features. The model interpolates the lip features constructed based on the opening and closing angles and combines audio and video in time series. Feature alignment and fusion and feature learning and classification are realized through the two-way LSTM SOFTMAX layer, and finally, end-to-end pronunciation error detection is realized through CTC. It is verified on the GRID audio and video corpus after phoneme conversion and the self-built multimodal test set. The experimental results show that the model has a higher false pronunciation recognition rate than the traditional single-modal acoustic error detection model. The increase in error detection rate is more obvious. Verification by the audio and video corpus with white noise was added, and the proposed model has better noise immunity than the traditional acoustic model.
... Common patterns of phone insertions, deletions and substitutions (accent errors) are collected from the literature for 10 L1s present in the BULATS, SELL and LeaP data sets used for error detection in this work (see §7.3). Error pattern are obtained from two aggregators [216,121] as well as from work specific to each L1, namely: French [105,244], Chinese [83,45,222,231,44,252], Vietnamese [261,203,237,8], Thai [53,143,124], Spanish [41,3,93,18,275], Russian [131,242,32], Dutch [82,287,63], Arabic [13,75,228], Polish [251,212,96] and German [30]. ...
Growing global demand for learning a second language (L2), particularly English, has led to considerable interest in automatic spoken language assessment, whether for use in computerassisted language learning (CALL) tools or for grading candidates for formal qualifications. This thesis presents research conducted into the automatic assessment of spontaneous nonnative English speech, with a view to be able to provide meaningful feedback to learners. One of the challenges in automatic spoken language assessment is giving candidates feedback on particular aspects, or views, of their spoken language proficiency, in addition to the overall holistic score normally provided. Another is detecting pronunciation and other types of errors at the word or utterance level and feeding them back to the learner in a useful way. It is usually difficult to obtain accurate training data with separate scores for different views and, as examiners are often trained to give holistic grades, single-view scores can suffer issues of consistency. Conversely, holistic scores are available for various standard assessment tasks such as Linguaskill. An investigation is thus conducted into whether assessment scores linked to particular views of the speaker’s ability can be obtained from systems trained using only holistic scores. End-to-end neural systems are designed with structures and forms of input tuned to single views, specifically each of pronunciation, rhythm, intonation and text. By training each system on large quantities of candidate data, individual-view information should be possible to extract. The relationships between the predictions of each system are evaluated to examine whether they are, in fact, extracting different information about the speaker. Three methods of combining the systems to predict holistic score are investigated, namely averaging their predictions and concatenating and attending over their intermediate representations. The combined graders are compared to each other and to baseline approaches. The tasks of error detection and error tendency diagnosis become particularly challenging when the speech in question is spontaneous and particularly given the challenges posed by the inconsistency of human annotation of pronunciation errors. An approach to these tasks is presented by distinguishing between lexical errors, wherein the speaker does not know how a particular word is pronounced, and accent errors, wherein the candidate’s speech exhibits consistent patterns of phone substitution, deletion and insertion. Three annotated corpora x of non-native English speech by speakers of multiple L1s are analysed, the consistency of human annotation investigated and a method presented for detecting individual accent and lexical errors and diagnosing accent error tendencies at the speaker level.
This paper reports on the role of technology in state-of-the-art pronunciation research and instruction, and makes concrete suggestions for future developments. The point of departure for this contribution is that the goal of second language (L2) pronunciation research and teaching should be enhanced comprehensibility and intelligibility as opposed to native-likeness. Three main areas are covered here. We begin with a presentation of advanced uses of pronunciation technology in research with a special focus on the expertise required to carry out even small-scale investigations. Next, we discuss the nature of data in pronunciation research, pointing to ways in which future work can build on advances in corpus research and crowdsourcing. Finally, we consider how these insights pave the way for researchers and developers working to create research-informed, computer-assisted pronunciation teaching resources. We conclude with predictions for future developments.
This book examines intelligibility in pronunciation teaching and its connections to L2 oral communication: the principles behind the concept of intelligibility; what research says about intelligibility for segmentals, consonant clusters, word stress, rhythm, prominence and final intonation; and what intelligibility research might look like in classroom contexts.
In this paper, we introduce L2-ARCTIC, a speech corpus of non-native English that is intended for research in voice conversion, accent conversion, and mispronunciation detection. This initial release includes recordings from ten non-native speakers of English whose first languages (L1s) are Hindi, Korean, Mandarin, Spanish, and Arabic, each L1 containing recordings from one male and one female speaker. Each speaker recorded approximately one hour of read speech from the Carnegie Mellon University ARCTIC prompts, from which we generated orthographic and forced-aligned phonetic transcriptions. In addition, we manually annotated 150 utterances per speaker to identify three types of mispronunciation errors: substitutions, deletions, and additions, making it a valuable resource not only for research in voice conversion and accent conversion but also in computer-assisted pronunciation training. The corpus is publicly accessible at https://psi.engr.tamu.edu/l2-arctic-corpus/.
Pronunciation training systems detect mispronunciations from language learner's speech and provide useful feedback. Mispronunciation detection systems can either be developed using Confidence Measures (CM) or using classifiers with Acoustic Phonetic Features (APF). This paper presents an APF based computer assisted pronunciation training (CAPT) system for most confusing Arabic phoneme pairs (/ط / vs ت / /)and (/ ح / vs / خ / or / هـ /) developed for subjects of Pakistani origin. A super-vector is formed based on APF consisting of Mel-frequency cepstral coefficients (MFCCs) along with its first and second derivative, energy, zero-cross, spectral features and pitch. A large dataset has been recorded from 200 speakers of Pakistani origin learning Arabic as their second language. Four different machine learning classifiers; Random Forest, Naïve Bayes, Ada-boost and K-NN have been used for mispronunciation detection. A comparison has been conducted between these classifiers and standard Goodness of Pronunciation (GOP) method. The results show that Random Forest outperforms all other methods by a significant margin.
The central purpose of this study is to illustrate how ESL instructors can take a principled approach to setting pronunciation instruction priorities for learners. Elicited speech samples from 30 adult English learners were analysed for suprasegmental and segmental pronunciation features. Guided by Levis’ (2005) intelligibility principle, results of the analysis led to recommended foci for pronunciation instruction. The study’s participants come from three distinct first language (L1) backgrounds (Mandarin Chinese, Colombian Spanish, and Slavic), reflecting the type of linguistic breadth found in typical ESL classrooms. It is recommended that problematic features observed in the speech of participants from all three L1s be addressed as a whole group, with each L1 group also receiving separate instruction for their specific difficulties. Finally, results of the speech analysis are compared with previously published material describing L1-specific pronunciation difficulties.
This paper presents the results of a project designed to functionally test the mutual intelligibility of spoken Maltese, Tunisian Arabic, and Benghazi Libyan Arabic. We compiled an audio-based intelligibility test consisting of three components: a word test where the respondents were asked to perform a semantic classification task with 11 semantic categories, a sentence test where the task was to provide a translation of a sentence into the respondent’s native language, and a text test where a short text was listened to twice and the respondents were asked to answer 8 multiple-choice questions. Data were collected from 24 respondents in Malta, Tunis, and Benghazi. It was found that there exists asymmetric mutual intelligibility between the two mainstream varieties of Maġribī Arabic and Maltese, with speakers of Tunisian and Libyan Arabic able to understand about 40 % of what is being said to them in Maltese, against about 30 % for speakers of Maltese exposed to either variety of Arabic. Additionally, it was found that Tunisian Arabic has the highest level of mutual intelligibility with either of the other two varieties. Combining the intelligibility scores with comparative linguistic data, we were able to sketch out the phonological variables involved in enabling and inhibiting mutual intelligibility for all three varieties of Arabic and set the stage for further research into the subject.
Many ESL/EFL students in Lebanon reach the university level with poor English in general and underdeveloped phonetic competence in particular. Phonetic competence is one of the abilities to communicate regarding pronunciation skills (Saz, Rodríguez, Lleida, Rodríguez, & Vaquero, 2011). Students in Lebanon acquire Arabic as their mother tongue. Some learn English as their second language (ESL) in schools where English is the medium of instruction. However, in French schools, students learn English as their foreign language (EFL) and study the content subject matters in French, the medium of instruction. It's commonly known among many educators in Lebanon that French educated students develop their English language skills and proper pronunciation more perfectly than do their English educated fellows in middle/low class schools. This phenomenon is widely observed; however, it is not based on research findings. This study aims to investigate whether there is a difference among French and English educated students in terms of English pronunciation after intervention of pronunciation learning strategies (PLS) such as using phonetic symbols and transcriptions, repeating after the teacher/others, and minimal pair drilling. 22 EFL and ESL university students took part in this study. A checklist of frequent common pronunciation mistakes was used to collect data. Ten sounds (segments) were identified in the pretest as common mistakes. The post-test took place five weeks later. Findings showed that PLS improved the pronunciation of both EFL and ESL learners. However, there was no statistical difference in the improvements between these two groups. Limitations and recommendations are provided in this study.
This study investigates the difficulties of English pronunciation encountered by Saudi secondary school learners when pronouncing English consonants. It also aims to shed light on the area of English consonant clusters system. The instruments used for collecting data and information included were questionnaires, classroom observations and document collections. The results show that the participants had difficulties to pronounce eleven consonant sounds. The results also demonstrate that a great number of the participants, unintentionally insert a vowel sound in English syllable to break up consonant clusters. This study provides some useful pedagogical implications to prevent and cure English pronunciation problems.
Mispronunciations of vowel and consonant (segmental) sounds are among the most frequent and identifiable types of difficulties in second language speech. This study examined the extent to which the
pronunciation of non-standard segmental sounds contributed to how often listeners noticed difficulties in understanding, and how the assessed spoken proficiency levels of nonnative English speakers were related to the native English speaking listeners’ understanding. Five linguistically-trained native American English listeners watched video-recorded teaching demonstrations of three Korean speakers of English whose oral proficiency in English had been rated at three different levels (low, intermediate, and advanced). The speakers were international teaching assistants (ITAs), or in preparation to be so. The listeners, using think-aloud techniques, paused whenever they had difficulty understanding. They then verbally described the nature of the difficulty they experienced, similar to the procedure used by Zielinski (2008). The findings showed that listeners stopped more frequently for the ITAs who had been rated as low and intermediate in oral ability than they did for the advanced speaker. In addition, the reasons for stopping varied according to the speakers’ levels. The findings of this study indicate that final consonants should be treated as an important carrier of grammatical, topical, and discoursal cues in academic talk. The findings also indicate that accuracy in some stressed vowel sounds is particularly important, and that they should be highlighted in teaching and mentoring ITAs.
Not all aspects of a language have equal importance for speakers or for learners. From the point of view of language description, functional load is a construct that attempts to establish quantifiable hierarchies of relevance among elements of a linguistic class. This paper makes use of analyses conducted on the 10-million-word spoken subcorpus of the British National Corpus in order to characterize what amounts to approximately 97% of the phonological forms and components heard and produced by fluent speakers in a range of contexts. Our aim is to provide segmental, sequential, and syllabic level rankings of spoken English that can serve as the basis for reference and subsequent work by language educators and researchers.
Research on the efficacy of second language (L2) pronunciation instruction has produced mixed results, despite reports of
significant improvement in many studies. Possible explanations for divergent outcomes include learner individual differences,
goals and foci of instruction, type and duration of instructional input, and assessment procedures. After identifying key
concepts, we survey 75 L2 pronunciation studies, particularly their methods and results. Despite a move towards emphasizing
speech intelligibility and comprehensibility, most research surveyed promoted native-like pronunciation as the target. Although
most studies entailed classroom instruction, many featured Computer Assisted Pronunciation Teaching (CAPT). Segmentals were
studied more often than suprasegmentals. The amount of instruction required to effect change was related to researchers’ goals;
interventions focusing on a single feature were generally shorter than those addressing more issues. Reading-aloud tasks were
the most common form of assessment; very few studies measured spontaneous speech. The attribution of improvement as a result
of instruction was compromised in some instances by lack of a control group. We summarize our findings, highlight limitations
of current research, and offer suggestions for future directions.
An important question in the L2 acquisition of phonemic distinctions is whether learners have to perceive contrasts before they can successfully produce them. In this paper, data on both the perception and production of the English /s/-/š/ contrast were elicited from native speakers of Korean, a language in which [s] and [š] are allophones of the same phoneme. General principles of phonology predict that, in acquiring this contrast, Korean learners will suppress the application of the allophonic rule relating these sounds across a morpheme boundary only if they can prevent the rule from applying within morphemes. This study examines these predictions in both production and perception of the /s/-/š/ contrast as a function of learner's training.
An analysis of the causes of misunderstandings and how they are dealt with from a corpus of 183 tokens of misunderstanding that occurred between speakers from different backgrounds.
Language experience systematically constrains perception of speech contrasts that deviate phonologically and/or phonetically from those of the listener's native language. These effects are most dramatic in adults, but begin to emerge in infancy and undergo further development through at least early childhood. The central question addressed here is: How do nonnative speech perception findings bear on phonological and phonetic aspects of second language (L2) perceptual learning? A frequent assumption has been that nonnative speech perception can also account for the relative difficulties that late learners have with specific L2 segments and contrasts. However, evaluation of this assumption must take into account the fact that models of nonnative speech perception such as the Perceptual Assimilation Model (PAM) have focused primarily on naive listeners, whereas models of L2 speech acquisition such as the Speech Learning Model (SLM) have focused on experienced listeners. This chapter probes the assumption that L2 perceptual learning is determined by nonnative speech perception principles, by considering the commonalities and complementarities between inexperienced listeners and those learning an L2, as viewed from PAM and SUA. Among the issues examined are how language learning may affect perception of phonetic vs. phonological information, how monolingual vs. multiple language experience may impact perception, and what these may imply for attunement of speech perception to changes in the listener's language environment.
It is argued that most if not all of the pronunciation problems encountered by Cantonese learners of English may be adequately accounted for by the contrastive differences discussed in this paper. The phonological differences between the two languages are examined, ranging from their phoneme inventories, the characteristics of the phonemes, the distributions of the phonemes, syllable structure, to the function of tones and their respective rhythmic patterns. At the segmental level, substitution by a related sound in the native language, deletion and epenthesis are by far the most common strategies Cantonese speakers employ when speaking or reading English. Pronunciation problems are also found at the suprasegmental level, that is, in connected speech and rhythm, resulting in the impression of a somewhat unnatural, “flat and boring” foreigner accent.
Computer-Assisted Language Learning (CALL) applications for improving the oral skills of low-proficient learners have to cope with non-native speech that is particularly challenging. Since unconstrained non-native ASR is still problematic, a possible solution is to elicit constrained responses from the learners. In this paper, we describe experiments aimed at selecting utterances from lists of responses. The first experiment on utterance selection indicates that the decoding process can be improved by optimizing the language model and the acoustic models, thus reducing the utterance error rate from 29–26% to 10–8%. Since giving feedback on incorrectly recognized utterances is confusing, we verify the correctness of the utterance before providing feedback. The results of the second experiment on utterance verification indicate that combining duration-related features with a likelihood ratio (LR) yield an equal error rate (EER) of 10.3%, which is significantly better than the EER for the other measures in isolation.
Contemporary views of adult pronunciation instruction emphasize the development of intelligible speech using empirically-validated pedagogical principles. Because learners typically have limited time for pronunciation work, instruction should be provided in a way that maximizes the use of the available opportunities. However, achievement of such a goal entails applying detailed knowledge of the phonetic learning process with due attention to the nature of differences that arise among learners, whether they share or do not share the same native language. In this longitudinal investigation, we examined productions of consonants and consonant clusters in English learners from two language backgrounds over a two-year period. Extensive between- and within-group variability was observed, with some targets produced very well at the outset, and others improving over time. The results argue against a common curriculum for learners. Instead, pronunciation instruction that focuses on individual learners' needs is called for. The findings are discussed in terms of strategies that might be used to develop effective and efficient pedagogical practices.
The claim that the best language-teaching materials are based on a contrast of the two competing linguistic systems has long been a popular one in language teaching. It exists in strong and weak versions, the strong one arising from evidence from the availability of some kind of metatheory of contrastive analysis and the weak from evidence from language interference. The strong version of the hypothesis is untenable and even the weak version creates difficulties for the linguist. Recent advances in linguistic theory have led some people to claim that the hypothesis is no longer useful in either the strong or the weak version. Such a claim is perhaps unwarranted, but a period of quiescence is probable for contrastive analysis itself.
The concept of functional load has been used by various writers in various linguistic fields and has consequently received differing definitions and methods of calculation. It has not, however, been applied to the teaching of pronunciation. This article examines several aspects of functional load in English that may be relevant for assessing the relative importance of segmental features of learners' speech. Implications for the use of pronunciation drill books are discussed.
The pronunciation of English syllable codas by second language (L2) learners of English, especially those whose native languages (L1s) do not have phonetic features similar to English, has received much attention in L2 research. Vietnamese, for example, does not have consonant clusters which are allowed in English in various word- positions. Vietnamese L2 learners of English have been found to have problems with pronouncing English consonant clusters, particularly those in word-final positions. The present research aimed to yield more insights into the interlanguage phonology of Vietnamese learners of English in this regard. Specifically, it was designed to identify (1) final clusters that Vietnamese L2 learners of English had difficulty producing accurately, (2) phonological processes that appeared to be at work when a final cluster was pronounced inaccurately, and (3) if task type had an effect on accuracy in final consonant production. The participants in this study were five Vietnamese students, enrolled in a U.S. university, whose performance on three tasks— reading a wordlist, reading a text, and semi-structured interview—was audio recorded and transcribed. The results showed that final consonant clusters which consisted of voiced obstruents (e.g., /bd/, /vz/ ) were more difficult than those containing voiceless obstruents. Final consonant clusters containing a liquid (e.g., /rt/, /lθ/) were also more problematic than those containing a nasal. Deletion was found to be the most common modification strategy and task type was found to have an effect on the articulation of final consonant clusters.
Languages convey information using several methods, and rely to di erent extents on di erent methods. The amount of reliance of a language on a method is termed the'functional load'of the method in the language. The term goes back to early Prague School days (Mathesius, 1929; Jakobson, 1931; Trubetzkoy, 1939), though then it was usually taken to refer only to the importance of phonemic contrasts, particularly binary oppositions.
This study investigated the relationship between experienced SPEAK Test raters' judgments of nonnative pronunciation and actual deviance in segmentals, prosody, and syllable structure. Sixty reading passage speech samples from SPEAK Test tapes of speakers from 11 language groups were rated impressionistically on pronunciation and later analyzed for deviance in segmentals, prosody, and syllable structure. The deviance found in each area of pronunciation was then correlated with the pronunciation ratings using Pearson correlations and multiple regression. An analysis of the 60 speakers showed that whereas deviance in segmentals, prosody, and syllable structure all showed a significant influence on the pronunciation ratings, the prosodic variable proved to have the strongest effect. When separate analyses were done on two language subgroups within the sample, prosody was always found to be significantly related to the global ratings, whereas this was not always true for the other variables investigated.
Although it is recognized that ESL students often need assistance to become more comprehensible speakers, their teachers usually have limited time to devote to pronunciation instruction. Research should help teachers set priorities for pronunciation teaching to address these students’ needs as efficiently as possible. Here we test the usefulness of the theoretical notion of functional load (FL) as a means of determining which consonant distinctions have the greatest impact on listeners’ perceptions of accentedness and comprehensibility. Although this principle has been proposed by several pronunciation experts, its predictions have not been empirically tested. Thirteen native English listeners judged 23 Cantonese-accented sentences that exhibited various combinations of high and low FL errors. The high FL errors had relatively large effects on both perceptual scales, while the low FL errors had only a minimal impact on comprehensibility. The only cumulative effects of errors seen in the data occurred with high FL errors in the judgments of accentedness. These results not only shed light on the distinction between accentedness and comprehensibility, but also suggest that the functional load principle can be effectively employed in guiding some aspects of pronunciation instruction.
This paper reports on an intervention study which investigated the effect of teaching English phonotactics upon Arabic speakers’ lexical segmentation of running speech in English. The study involved a native English-speaking group (N = 12), a non-native control group (N = 20); and a non-native experimental group (N = 20). Each group was pre-tested using a Word Spotting Task which investigated the extent to which illegal consonant clusters in English and Arabic supported the lexical segmentation of English. The non-native groups were post-tested with the same task after 8 weeks, during which the experimental group was given a treatment consisting of explicit teaching of relevant English phonotactic constraints. Post-test results showed significant gains in the segmentation ability of the experimental group.
In this study I investigate the impact of different characteristics of the L2 speech signal on the intelligibility of L2 speakers of English to native listeners. Three native listeners were observed and questioned as they orthographically transcribed utterances taken from connected conversational speech produced by three L2 speakers from different L1 backgrounds (Korean, Mandarin, and Vietnamese). Where the listeners experienced difficulty in identifying the speaker’s intended words, the features of the speech signal they relied on were identified, and links were made between the difficulties they experienced and the way in which the features in question varied from standard English phonology. Regardless of the speaker they were listening to, when attempting to identify the speaker’s intended words, all three listeners relied heavily and consistently on the speaker’s syllable stress pattern (the number and pattern of strong and weak syllables), and more consistently on segments in strong syllables than those in weak syllables. Both non-standard syllable stress patterns and non-standard segments misled them into wrongly identifying a speaker’s intended words, but for all three speakers, it was non-standard segments in strong syllables that misled those most often. These findings have implications for L2 listeners wanting to improve their English listening skills, as well as for L2 speakers wanting to improve their intelligibility.
Frequency counts are a measure of how much use a language makes of a linguistic unit, such as a phoneme or word. However, what is often important is not the units themselves, but the contrasts between them. A measure is therefore needed for how much use a language makes of a contrast, i.e. the functional load (FL) of the contrast. We generalize previous work in linguistics and speech recognition and propose a family of measures for the FL of several phonological contrasts, including phonemic oppositions, distinctive features, suprasegmentals, and phonological rules. We then test it for robustness to changes of corpora. Finally, we provide examples in Cantonese, Dutch, English, German and Mandarin, in the context of historical linguistics, language acquisition and speech recognition. More information can be found at http://dinoj.info/research/fload
Methods of teaching English to Arab students
- N Al-Mutawa
- T Kailani
Al-Mutawa, N., & Kailani, T. (1989). Methods of teaching English to Arab students. Harlow:
Longman.
Improving English pronunciation among Arabic EFL school-age students using minimal pairs
- A K Altamimi
Altamimi, A.K. (2015). Improving English pronunciation among Arabic EFL school-age students
using minimal pairs. Unpublished masters thesis, SUNY Fredonia, Fredonia, NY, USA.
Corpus analysis of spoken discourse: Research findings, prospects, implications for teaching
- D Biber
Biber, D. (2019). Corpus analysis of spoken discourse: Research findings, prospects, implications for teaching. Plenary speech given at Pronunciation in Second Language Learning and
Teaching conference, September 2019, Northern Arizona University, Flagstaff, AZ, USA.
Phonetics and the teaching of pronunciation
- J C Catford
Catford, J.C. (1987). Phonetics and the teaching of pronunciation. In Morley, J. (Ed.), Current
perspectives on pronunciation (pp. 87-100). Alexandria, VA: TESOL Press.
Teaching pronunciation: A course book and reference guide
- M Celce-Murcia
- D M Brinton
- J M Goodwin
Celce-Murcia, M., Brinton, D.M., & Goodwin, J.M. (2010). Teaching pronunciation: A course
book and reference guide. Cambridge: Cambridge University Press.
The quantification of functional load: A linguistic problem
- C F Hockett
Hockett, C.F. (1966). The quantification of functional load: A linguistic problem. ERIC. Available
at: https://files.eric.ed.gov/fulltext/ED011649.pdf (accessed May 2020).
Errors in English among Arabic speakers: Analysis and remedy
- N Kharma
- A Hajjaj
Kharma, N., & Hajjaj, A. (1989). Errors in English among Arabic speakers: Analysis and remedy.
Harlow: Longman.
English pronunciation for Arabic speakers
- T F Mitchell
- S El-Hassan
Mitchell, T.F., & El-Hassan, S. (1989). English pronunciation for Arabic speakers. Harlow:
Longman.
How well can we predict second language learners’ pronunciation difficulties?
- M J Munro
Munro, M.J. (2018). How well can we predict second language learners' pronunciation difficulties? CATESOL Journal, 30, 267-281.
Pronunciation and intelligibility of Turkish native speakers learning English
- T Uzun
Uzun, T. (2019). Pronunciation and intelligibility of Turkish native speakers learning English.
Unpublished PhD thesis, Ankara University, Ankara, Turkey.
The intelligibility cocktail: An interaction between speaker and listener ingredients
- B Zielinski
Zielinski, B. (2006). The intelligibility cocktail: An interaction between speaker and listener ingredients. Unpublished PhD thesis, MacQuarie University, Sydney, Australia.
Developmental and transfer factors in English pronunciation of Korean elementary school and college students
- G No