ALPHABETIC LITERACY AND PRACTICAL PHONETICS TEACHING: SOME PRELIMINARY CONNECTIONS
Ian Crookston
Speech Therapy
Leeds Metropolitan University

It is the standard view in the psychology of reading that literacy in an alphabetic writing system involves a conscious system of phonological representation (Adams 1990, Perfetti 1994). Presumably, then, to teach phonemic transcription of the student's native language is to enrich the phonological element of the student's literacy. The question then arises as to the exact extent of this element of literacy: what are we building on?

The beginning transcriptions of 20 first-year undergraduates in Speech & Language Therapy will be analysed for areas of difficulty. Two predictions will be tested.

Firstly, it is predicted that some common difficulties will be found to come from phonemes which are not represented in orthography in one sense or another; or whose orthographic representation is not transparent. It is further predicted that a common form of literacy, as evidenced by beginning students of phonemic transcription, is not distinctive-feature-based.

The implications of this approach for the understanding of four categories of learner will be discussed.

Adams M 1990 Beginning to Read MIT Press, Cambridge, Mass
Perfetti C 1994 "Psycholinguistics and reading ability" in M Gernsbacher ed Handbook of Psycholinguistics Academic Press, San Diego, 849-894

THE RELATIVE PERCEPTUAL ATTENTION TO ACOUSTIC CUES IN THE L2 LEARNING OF NEW CONTRASTS: FINDINGS AND APPLICATIONS
Paola Escudero
University of Reading
(PDF file not available)
L2 (second language) speech studies suggest that adult Spanish speakers of English, unlike native speakers, may rely more on durational than on spectral cues when identifying and discriminating the sounds involved in tense/lax vowel contrasts. However, the knowledge attained and the different patterns/stages in the L2 development have not been fully addressed.

Data from a perception experiment reveal that adult Spanish speakers not only use durational information more than L1 speakers but also that they can perceive Scottish English /i/-/I/ purely or more reliably (than using spectral cues) on this basis. Firstly, it is suggested that this may constitute evidence for L2 "category formation" or "phonological lexicalisation" (/i/ and /I/) based on a different weighting of the acoustic information involved. Secondly, it will be argued that the patterns found throughout the L2 speakers' performance might reflect a stage-like development in the learning of new contrasts. Furthermore, it will be suggested that the starting point for the L2 perception of new contrasts together with the possibilities for ultimate attainment may depend on the acoustics of both the target language and the L1 sounds.

I will take up on the implications and applications (i.e. speech training) of these conclusions as well as some current/further research toward the end of the presentation.

TEACHING ACCENTUAL FOCUS TO CATALAN STUDENTS OF ENGLISH
Eva Estebas-Vilaplana
Department of Phonetics and Linguistics
University College London

Catalan has traditionally been classified as a (-plastic) language, that is, a language that uses syntactic strategies to achieve the accent/focus association. This basically involves the movement of the focussed word to an accent-bearing location (i.e. sentence-final position). This behaviour contrasts with (+plastic) languages, such as English, where focus is signalled by alternating the intonational pattern of the sentence (i.e. the accent shifts over the focussed word), while the syntactic structure is unaffected. This classification suggests that Catalan students of English need to learn a completely new strategy to signal focus. However, recent research on Romance languages has shown that the (+/- plastic) parametric division of languages is not so clear-cut. Thus, an experiment was designed to investigate whether accentual focus can be used in Catalan (both produced and recognised) and whether it is an accepted strategy. Three tests were carried out: a production test (34 speakers), a perception test (28 speakers) and an acceptability test (28 speakers). The production test consisted of both reading and semi-spontaneous data. Two variables were included in all tests: focus triggered by contrast and by identification. The results showed that Catalan speakers can both produce and recognise accentual focus mainly with a contrastive meaning. Similarly, accentual strategies were more accepted in contrast contexts than in identification contexts. The immediate implication of these findings is that Catalan students of English do not need to learn a new strategy to signal focus but only broaden the domains of its application.

AUTOMATIC FEEDBACK ON PHONETIC TRANSCRIPTION
María Luisa García Lecumberri (University of the Basque Country)
Martin Cooke (University of Sheffield)
John Maidment (UCL)

Phonemic transcription is one of the requirements of many phonetics and phonology and even general linguistics courses at university level . Transcription of one's native language is a useful mechanism for raising awareness of the sound system and its behaviour in connected speech, making explicit the speaker's competence. Transcription is also used in language courses for non-linguistics students.

The provision of early, detailed and relevant feedback on learners' transcription attempts is a valuable adjunct to formal classes, but is labour-intensive at present. Consequently, a tool which provides learners
with the means to input transcriptions, and then to receive detailed, multi-faceted, feedback on their efforts, has been produced. The tool operates by comparing a learner's transcription with a model provided by the tutor. It employs string alignment algorithms, similar to those used in the assessment of speech technology, modified to accommodate alternatives and optional symbols.

The tool has been developed for British English (RP) but has been designed with ease of internationalisation in mind via the use of XML for all external data. It is anticipated that examples of such localisation, together with extensions to support dictation exercises will be implemented in the coming months.

The tool will soon be available as a Java applet with associated database on the web.

ANALYTICAL TRAINING FOR THE TEACHING OF EFL IN PRIMARY SCHOOL
Encarna Gayoso
Mercedes Blanco
Marisol Carrillo
University of Murcia

This paper describes the results obtained with a modified training programme carried out in 1997/98 and based on a pilot programme run in 1996/97, the results of which were presented at the inaugural meeting of PTLC in 1999. Our intention was to study the effectiveness of providing phonological/analytical training in English as L2 and to compare the results with those of non-analytical training.

The novelty of the training lies in the teaching of pronunciation through phono- and metaphonological awareness by, among other things, analysing the articulatory gestures used to make sounds.

The theoretical basis of the ongoing study is the relationship between metaphonological development and the acquisition of reading and writing skills in alphabetic languages.

The pilot experiment was carried out in 1996/97 in a state primary school in Murcia (Spain) and the modified training programme, the results of which we describe, was carried out in a different primary school in Murcia with the main differences being:

- the intrasyllabic level (onset and rime) was given more importance in the second study.

- the number of subjects increased to 28 (all the pupils registered in the second year of Primary Education in that school.

- both the familiarisation with English (2 months) and the training (4 months) were carried out as part of the class timetable and not as an extracurricular activity as in 96/97.

After familiarisation with the language, the pupils were given a test to measure their initial competence in phonological production (word repetition, vocabulary, phonological awareness in L2) in order to create two homogeneous groups (experimental and control).

The two groups were then taught the same lexicon of 90 words in two distinct programmes, one analytical and the other non-analytical.

After four months of teaching the subjects were again evaluated. The results showed that the experimental group performed better than the control group in reading, writing and intrasyllabic segmentation but not in vocabulary.

A MULTIMEDIA APPROACH TO SUPRASEGMENTALS: USING A CD-ROM FOR ENGLISH INTONATION TEACHING
Gunther Kaltenboeck
Department of English
University of Vienna

In this paper I would like to demonstrate the prototype of a CD-ROM for teaching English intonation to EFL (English as a Foreign Language) students and point out some of the advantages and limitations of this medium for pronunciation teaching. One of the key characteristics of the CD-ROM is its emphasis on learner autonomy. Not only does it allow learners to work independently of a teacher or a classroom and at their own pace, but it also gives students considerable control over the sequence of exercises, encouraging them to choose their own path through the CD-ROM. This flexibility in the CD-ROM structure opens up the possibility of accessing intonation in fundamentally different ways, which allows to accommodate different learner types and learning styles. Thus, the user is given the choice, for instance, of whether s/he wants to start with a more theoretical explanation of a prosodic concept and then move on to its practical application in a specific speech situation, or vice versa, moving from contextualised examples to systematic exercises.

THE ROLE OF THE MOTHER TONGUE UPON THE ACQUISITION OF ENGLISH TONALITY AND TONICITY RULES
Smiljana Komar
Dept. of English, Faculty of Arts, University of Ljubljana

The paper examines the differences and similarities between the English and Slovene tonality and tonicity rules, and discusses the problems that Slovene students of English might face when reading and/or speaking in English.

Our expectations regarding the problems of tonality and tonicity in English were tested with the first, second and fourth year undergraduate students at our department. Students were divided in two groups and were tested in reading of an unknown English text as well as its Slovene translation. The results in the distribution of intonation units and the choice of nuclear syllables in the English and Slovene readings were compared and analysed.

The results showed a relatively high percent of overlapping between the English and Slovene readings in both, the distribution of the text into intonation units as well as the nucleus placement. At the same time it also showed that the awareness of the differences between English and Slovene tonality and tonicity is higher with the fourth year students than with the first or second year ones. In other words, making students aware of these differences is useful and productive and should continue to be one of the main goals in the teaching of English prosody to Slovene students of English.

HOW WE ARE STRESSED?! PHONETIC CORRELATES AND STRESS PLACEMENT IN SINGAPOREAN ENGLISH
Lisa Lim
Tan Ying Ying
National University of Singapore

Most work on Singaporean English (SE) stress has assumed its phonetic correlates to be comparable to those in Standard British English. Low (1998), for example, investigating lexical stress placement, assumes the higher pitched syllable to be stressed, even though pitch has been noted to be an unreliable cue for SE stress (Yeow 1987). Our research shows that not only do the phonetic cues for SE stress differ from StdBrE, but also that Chinese, Indian and Malay Singaporeans each seem to use different cues for stressed syllables, compared to unstressed syllables (Tan ms), and syllables/words which should receive emphatic stress are stressed no differently from those with neutral stress. Additionally, young Singaporeans do not stress new or contrastive information (cf. Goh 2000), although older generation SE speakers do. The phonetic realisation of stress in different languages or in different varieties of a language varies widely, and we feel that the phonetic correlates of stress in SE have to established first before any reasonable investigation of other phenomena in SE stress can be conducted. An awareness of this as well as of stress placement in SE are vital in the teaching and learning of prosody as well as in effective communication.

References
Goh, C.C.M. 2000. A discourse approach to the description of intonation in Singapore English. In A. Brown, D. Deterding & E.L. Low (eds.). The English Language in Singapore: Research on Pronunciation. Singapore: SAAL. 35-45.
Low, E.L. 1998. Prosodic prominence in Singapore English. PhD dissertation. University of Cambridge.
Tan, Y.Y. ms. Acoustic correlates of stress in the ethnic sub-varieties of Singapore English. Department of English Language & Literature, National University of Singapore.
Yeow, K.L. 1987. Stress, rhythm and intonation in educated Singapore English: An auditory and instrumental study. MA dissertation. Department of English Language and Literature, National University of Singapore.

TEACHING PHONETICS WITH A MULTIMODAL INTERNET SITE
François PELLEGRINO
Christian FRESSARD
Gilbert PUECH
Université Lumière Lyon 2

Teaching phonetics without audiovisual materials is frustating for both
teachers and learners! Fortunately, recent advances in computer science,
provide user-friendly, efficient ways to share knowledge and to help
students to assimilate complex notions.
This paper describes a Web server developped at the Language Science
Department of the University of Lyon (France). It is dedicated to the
teaching of Phonetics for second and third year undergraduates.
The main purpose is to enable students to overcome their common lack of
fundamentals on wave propagation, spectral analysis, etc [1].
The server contains multimedia documents and exercices which were
developped using Flash [2] and Shockwave [3] technologies for the
graphical part and java and Tcl/Tk (with the Snack toolkit [4]) for the
sounds.
The multimodal capacities of Snack give the student the possibility to
process in real time a sound recorded via the web: this way, each
student may record his own voice and see in real time the resulting
spectrogram, the pitch contour, etc. as he may do it using a standard
speech analysis software.
The Web server design is still in progress, and it will be available
this winter.

References :
[1] J. Clark and C. Yallop, "An Introduction to Phonetics & Phonology",
Basil Blackwell, Oxford and Cambridge Ed., 1990
[2] http://www.macromedia.com/software/flash/, visited in 10/00
[3] http://www.macromedia.com/software/director/, visited in 10/00
[4] http://www.speech.kth.se/snack/, visited in 10/00

PHON2: PHONETICS BEYOND THE BASICS
Peter Roach, Simon Arnfield, Pierre Busuttil, Snezhina Dimitrova, Tanya Kostadinova, Michael Parsons, Valerie Roques and Monique Verrac

The paper describes a "European Module" project funded by the European Union through ERASMUS to provide intermediate-level phonetics tuition via the internet. Our work has three fundamental principles:

1. We assume that students using PHON2 will have already encountered phonetics in the context of learning the pronunciation of a foreign language, and are now embarking on a more advanced and detailed study of the subject.
2. There is a large amount of useful teaching material already available on the web, but students need careful guidance in using it for study purposes.
3. At all points in the module, students should be provided with exercise material which includes acoustic analysis.

The current state of the module can be seen on the website, http://www.linguistics.reading.ac.uk/phon2

THE ACQUISITION OF ENGLISH PLOSIVES BY CHINESE LEARNERS

Joanna Radwanska-Williams
Josephine P S Yam
The Chinese University of Hong Kong


The aim of the paper is to examine the acquisition of English plosives in syllable final position by Cantonese and Mandarin speaking EFL learners. From recording the pronunciation of a list of English words with syllable final plosives of 6 Mandarin and 6 Cantonese EFL learners, we have found the following phenomena in their pronunciation:
1.) Schwa epenthesis after syllable final English plosives
2.) Deletion of the syllable final plosives
3.) Production of unreleased final plosives.

It is also found that Mandarin EFL learners will tend to display epenthesis and deletion while Cantonese EFL learners produce unreleased plosives and deletion. We try to apply Optimality Theory (OT) to explain these interlanguage phenomena since it can handle the dynamic environment of interlangauge. OT argues that the phonological differences between languages are the results of the differences in the ranking of universal constraints. Thus, the phenomena produced by Cantonese and Mandarin EFL learners can be viewed as the differences in the ranking of constraints.

All of these phenomena (epenthesis, deletion, unreleased plosives) affect the intelligibility of English pronounced with a "Chinese accent", because they violate the OT faithfulness constraint, which is highly ranked in English. The restructuring of constraint rankings in the interlanguage is possible over a long period of time; however, it seems difficult to achieve over the short period of time typical of a classroom setting. We recommend raising the learners' language awareness of phonological phenomena to equip them for possible restructuring over life-long learning.

SPECO: COMPUTER-BASED PHONETIC TRAINING FOR CHILDREN
Anna Sfakianaki, Peter Roach, University of Reading,
Klara Vicsi, Ferenc Csatari, Technical University of Budapest
Anne-Marie Öster, KTH
Zdravko Kacic, University of Maribor, Slovenia
Peter Barczikay,RCS - Robot Control Software,


The SPECO Project (funded by the EU through the INCO-COPERNICUS program) is developing a workstation that provides real-time visual display of acoustic information for children in need of assistance with various aspects of speech production. Sophisticated signal processing technology incorporating an auditory model is employed to overcome normalisation problems resulting from inter-speaker differences and tempo variations. This is accomplished through the analysis of the speech signal by approximately the same time, frequency and intensity resolution available to the human peripheral auditory system. The languages currently implemented are Hungarian, German, English, Slovenian and Swedish. SPECO can be adapted to any European language through a general language-independent database editor and measuring system. An important feature of the system is that the decision as to what constitutes a "normal" pronunciation in each language is made with reference to a database of recordings of children's speech rated for acceptability by expert listeners. The exercise material is carefully graded, and includes practice on segments and on rhythm and intonation. The graphics are designed to be interesting and comprehensible to young children. Preliminary trials of the system have proved encouraging, and the final phase of the project will embody more extensive investigation of the system's capabilities.

USING A SPEECH RECOGNITION PROGRAM IN TEACHING ENGLISH PRONUNCIATION
Rastislav Sustarsic
University of Ljubljana

In addition to the existing computer programs that have been designed specifically for foreign language pronunciation (or intonation) training, a number of software packages are available today that can be used in phonetics classes, although they have not been developed for this particular purpose. I have been particularly intrigued by speech synthesis and speech recognition programs, and the present paper will focus on the latter.

One of the several commercial programmes available for English (IBM ViaVoice) will be analyzed from the point of view of its general usability and weaknesses, and then evaluated with regard to its applicability in English pronunciation classes. In the process of working with this particular program (and probably any other speech recognition program), we find that the 'errors' occurring in the process of interpretation of the speaker's (student's) intended message can be exploited both by way of discussing the various causes of confusion (inadequate pronunciation, homophony, phonetic processes on the word boundary), as well as by the subsequent effort of the student to achieve the desired result (the correct orthographic version of the spoken message).

JAPANESE EFL LEARNERS' WEAK POINTS IN ENGLISH INTONATION
Masaki Taniguchi
Chikushi Jogakuen University

This paper is based on research dedicated to the improvement of teaching and learning English intonation for Japanese EFL learners. Through experiments of fundamental frequencies (henceforth Fx) using the Laryngograph Processor, it outlines three of the weak points Japanese EFL
learners have in English intonation: (1) Narrow pitch range, (2) Nucleus "misplacement," and
(3) Interference of Japanese pitch accent patterns in their English intonation.

INTONATION OF ENGLISH IN THE PROCESS OF SECOND-LANGUAGE ACQUISITION.
Renáta Timková
Prešov University

The presentation is going to concentrate on one of the very important suprasegmental feature of language - intonation. The special focus will be on intonation of English language produced by Slovak learners of English. The main intention of the presentation is to find areas in this non-segmental phenomenon which make remarkable differences between intonation realization of Slovak learners of English and native speakers of English. It will concentrate on intonation of English declarative sentences which are analyzed in the process of reading with Slovak learners of English. The presentation is supported by instrumental measurements based on Multi - speech computer program speech analysis. The data is based on vocalic nuclei segmentation of each syllable - 2260 syllables altogether, and it includes formal readings of texts of the Slovak students of English and the native speakers of English. All the measurements have been carried out at the Phonetic Laboratory of the Department of Slovak language at the Prešov University, Slovakia. The presentation will point out the main differences in intonation contour of Slovak non-native speakers of English (NNS) and native speakers of English (NS). It will also try to outline the reasons of the differences through the process of intonation analysis. It will concentrate on the main problems in intonation-group boundary indentification, nucleus placement within an intonation-group, pitch fluctuations and nuclear tones identification in the utterance of NNS compared with NS. The analysis follows the British approach and adopts the terminology of A.Cruttenden. The presentation is covered by the problem of language interference, on the level of intonation, from L1 (language 1 - Slovak language) to L2 (language 2 - English language). It tries to prove the pressure of the primary intonation system (L1) on the secondary intonation system (L2) in the process of intonation awareness of Slovak learners of English. The choice of the material for analysis is that the value of texts read aloud is quite simply that they yield speech which can be readily compared from one speaker to another. The research on intonation awareness of Slovak English learners is at its initial step and in the future it will proceed to analysis of spontaneous speech.

CONTEXT EFFECTS OF PHONEMIC CATEGORISATION IN FOREIGN LANGUAGES - IMPLICATIONS FOR LANGUAGE TEACHING
Eivind Nessa Torgersen
University of Reading

Various studies have shown that L2 learners have problems in successfully producing and perceiving certain speech sounds, e.g. stop consonants, in languages different from their own. In addition, there seems to be a relation between those two abilities.
This presentation will look at how context effects, which are helpful in categorising phonemes in L1, may be able to work the same way in L2. Instead of learning and teaching speech sounds in isolation and in individual words, will the presence of a semantic context produce an improvement?
The paper will report the progress of an ongoing study of online phonemic categorisation in L1 and L2 with English speakers.

INNOVATIVE APPROACHES TO THE TEACHING OF PRACTICAL PHONETICS
Magdalena Wrembel
Teacher Training College,
Adam Mickiewicz University

New developments in pronunciation pedagogy have been affected by clear influences from other disciplines such as psychology, neuro-linguistics, drama and technology. This paper presents an overview of innovative approaches to the teaching of practical phonetics and points to their common characteristics. It is suggested that today's pronunciation curricula incorporate, beside traditional phonetic instruction and exposure, some new strategies adopted from multidisciplinary and holistic approaches to second language learning.

A brain-friendly approach to teaching practical phonetics, which manifests itself in the use of multisensory modes, i.e. auditory, visual, tactile and kinaesthetic reinforcements, is demonstrated to enhance acquisition by appealing to different learning styles. Drama techniques such as mimicry, shadowing or voice modulation are shown to be very effective in enabling students to gain a better control over their articulation and overcome fossilised pronunciation. The Neuro-Linguistic Programming (NLP) perspective is also advocated as it deals efficiently with affective and emotional factors related to learning pronunciation and facilitates an accurate production of L2 sounds through such techniques as relaxation, suggestion, visualisation and others. Finally, the results of a preference questionnaire conducted among students of the Teacher Training College are presented.

It is hoped that this paper will provide teachers of foreign language pronunciation with practical insights into a variety of innovative techniques and resources, help them expand the repertoire of traditional classroom practices and, consequently, enhance pronunciation instruction.