A case study on the effectiveness of elsa on learning english pronunciation with first year english language teaching major students at hanoi pedagogical university 2

HANOI PEDAGOGICAL UNIVERSITY 2 FACULTY OF FOREIGN LANGUAGES DO HOANG ANH A CASE STUDY ON THE EFFECTIVENESS OF ELSA ON LEARNING ENGLISH PRONUNCIATION WITH FIRST YEAR ENGLISH LANGUAGE TEA

HANOI PEDAGOGICAL UNIVERSITY 2 FACULTY OF FOREIGN LANGUAGES

DO HOANG ANH

A CASE STUDY ON THE EFFECTIVENESS OF ELSA ON LEARNING ENGLISH PRONUNCIATION WITH FIRST YEAR ENGLISH LANGUAGE TEACHING MAJOR STUDENTS AT HANOI PEDAGOGICAL UNIVERSITY 2

(SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT OF THE

DEGREE OF BACHELOR OF ARTS OF ENGLISH)

SUPERVISOR: DO THI THANH DUNG, M.A

HANOI, 2018

ACKNOWLEDGMENTS

First and foremost, I would like to express my deepest gratitude to my research supervisors, Mrs Do Thi Thanh Dung, M.A, without her guidance and whole-hearted instruction in every step throughout the process, this study would have never been accomplished As a novice researcher, I had made a lot of mistakes

in the process of writing the research paper but Mrs Dung was kind enough to work nights and days to help me completing the study I would like to give my sincerest thanks to her not only for being my research supervisor but for her support during

my whole four years at Hanoi Pedagogical University 2

I would also want to thank my lectures at Faculty of Foreign Languages for helping me and giving me the constructive comments of my thesis, their advices were invaluable to the results of this research

I wish to acknowledge the assistance provided by the students of K43A- English Language Teaching at Hanoi Pedagogical University 2, without their participation, I could not be able to conduct this study

Most importantly, I am truly grateful to my beloved family and my friends who were always there to help me get back in track, encourage me and give me power to overcome any challenges in my life

Once again, thank you all for your dedication!

ABSTRACT

This study aims to investigate students’ attitudes towards the implementation

of ELSA and to examine the effectiveness of ELSA in learning English pronunciation skills on a group of 10 students from K43A – English Language Teaching, Hanoi Pedagogical University 2 (HPU2) The data of this qualitative and quantitative study was collected from four different sources: Survey questionnaire, Interview, ELSA Assessment Test score and the Student Oral Language Observation Matrix (SOLOM) score From the results of the survey questionnaire and the interview, the researcher concluded that all of the participants had positive attitudes towards the application of ELSA as a supporting tool for self-study, the majority of them wanted to continue using ELSA to improve their pronunciation skills in the future Based on the results of the pre and post ELSA Assessment Test

as well six-week class observation, the implementation ELSA was demonstrated that it was effective in both improving students’ pronunciation skills and boosting their confidence and increasing their speaking fluency

STATE OF AUTHORSHIP Title: A Case Study on the Effectiveness of Elsa on Learning English

Pronunciation with First Year English Language Teaching Major Students at Hanoi Pedagogical University 2

I certify that no part of this research has been copied from any other person’s work without acknowledgements and the study is originally written by me under the guidance of my supervisor

Date submitted: 9 May, 2018

LIST OF TABLES

Table 1 English Consonant Phoneme

Table 2 The schedule of implementing ELSA in 6 weeks

Table 3 The SOLOM Assessment Score

Table 4 The comparison of ELSA score and SOLOM score from the participants Table 5 The SOLOM Assessment results of Week 1

Table 6 The SOLOM Assessment results of Week 2

Table 7 The SOLOM Assessment results of Week 3

Table 8 The SOLOM Assessment results of Week 4

Table 9 The SOLOM Assessment results of Week 5

Table 10 The comparison of ELSA score and SOLOM score from the participants Table 11 The SOLOM Assessment results of Week 6

LIST OF FIGURES

Figure 1 Features of pronunciation

Figure 2 English monothongs

Figure 3 The scores of ELSA Pre-Test and ELSA Post-Test

Figure 4 The scores of the Pre-SOLOM Pronunciation Score and

the Post-SOLOM Pronunciation Score

Figure 5 SOLOM Score in Week 1 and SOLOM Score in Week 6

Figure 6 Students’ methods of learning English pronunciation skills

Figure 7 Amount of time per week that students spent on learning English

pronunciation

Figure 8 Student’s difficulties in learning English pronunciation

Figure 9 Student’s difficulties in learning English consonants

Figure 10 Student’s difficulties in learning English vowels

Figure 11 Students’ opinions towards the usefulness of ELSA in pronouncing

Figure 14 Sounds that ELSA may not recognize properly

Figure 15 Students’ attitude towards the helpfulness of ELSA

Figure 16 Students’ biggest improvements in pronunciation after using ELSA

Figure 17 Students’ attitudes towards applying ELSA in a real curriculum

Figure 18 Students’ opinions on biggest drawbacks of the application of ELSA

Figure 19 Students’ recommendations for the implementation of ELSA

LIST OF ABBREVIATIONS

Hanoi Pedagogical University 2: HPU2

Automatic speech recognition: ASR

Computer Assisted Language Learning: CALL

Computer Assisted Pronunciation Training: CAPT

Hidden Markov Model: HMM

English Language Speech Assistant: ELSA

Student Oral Language Observation Matrix: SOLOM

Operating System: OS

International Phonetic Alphabet : IPA

TABLE OF CONTENTS

PART 1: INTRODUCTION 1

I RATIONALE OF THE STUDY 1

II AIMS OF THE STUDY 1

III RESEARCH QUESTIONS 2

IV THE SCOPE OF THE STUDY 2

V METHODS OF THE STUDY 2

VI SIGNIFICANCE OF THE STUDY 2

VII DESIGN OF THE STUDY 3

PART 2: DEVELOPMENT 4

CHAPTER I: LITERATURE REVIEW 4

A Pronunciation 4

1 Definition of pronunciation 4

1.1 English vowel sounds 5

1.1.1 Monophthongs 5

1.1.2 Diphthongs 6

1.1.3 Triphthongs 6

1.2 English consonant sounds 6

1.2.1 English consonants classification 7

1.2.1.1 Voicing 7

1.2.1.2 Manner of articulation 7

1.2.1.3 Places of articulation 8

B Speech recognition technology and ELSA 8

1 Definition of speech recognition 8

2 An introduction to ELSA Speak 9

3 The use of speech recognition in learning pronunciation 10

CHAPTER II: METHODOLOGY 12

1 Research method 12

1.1 Case study 12

1.2 Research procedure 12

1.2.1 Describing background 13

1.2.2 Identifying problems 13

1.2.3 Vision & objective(s) 13

1.2.4 Proposing solutions 14

1.2.5 Implementing case study 15

1.2.6 Results and evaluation 16

2 Data collection instruments 16

2.1 Survey questionnaire 16

2.1.1 Reasons for using questionnaire 16

2.1.2 Description of survey questionnaire 16

2.2 Classroom observation 18

2.2.1 Reasons for using classroom observation 18

2.2.2 Procedure of classroom observation 18

2.3 Interviews 19

2.3.1 Reasons for using interview 19

2.3.2 Description of interview 19

3 Research setting 19

3.1 Participants 20

4 Data analysis description 20

4.1 The ELSA Assessment Test 20

4.2 Classroom observation 20

4.3 Survey questionnaire 20

4.3.1 The pre-survey questionnaire 21

4.3.2 The post-survey questionnaire 21

CHAPTER 3: FINDINGS AND DISCUSSION 22

FINDINGS 22

1 The results of using ELSA in learning English pronunciation skills in 6 weeks 22

1.1 Description 22

1.1.1 Week 1 (19/03/2018 – 25/03/2018) 22

1.1.2 Week 2 (26/03/2018 – 01/04/2018) 23

1.1.3 Week 3 (02/04/2018 – 08/04/2018) 24

I.1.4 Week 4 (09/04/2018 – 015/04/2018) 24

1.1.5 Week 5 (16/04/2018 – 22/04/2018) 25

1.1.6 Week 6 (23/04/2018 – 29/04/2018) 25

1.1.7 The comparison of the Pre and Post ELSA Assessment Test and Pre and Post SOLOM Assessment 26

2 Students’ attitude towards using ELSA in learning English pronunciation skills 29

2.1 Results from survey questionnaire 29

2.1.1 Results the pre-survey questionnaire 29

2.1.1.1 Students’ attitude towards the importance of English pronunciation skills 29

2.1.1.2 Students’ interest in English pronunciation skills 30

2.1.1.3 Students’ methods of learning English pronunciation skills 30

2.1.1.4 The number of hours per week students spent on learning English pronunciation skills 31

2.1.1.5 Student’s difficulties in learning English pronunciation skills 32

2.1.1.6 Students’ experience in learning English pronunciation with ELSA 33

2.1.2 Results from the post-survey questionnaire 35

2.1.2.1 Students’ opinions towards the usefulness of ELSA in learning English pronunciation 35

2.1.2.2 Students’ opinions towards the implementation of ELSA in learning English pronunciation 37

2.1.2.3 Challenges that students faced in the process of implementing ELSA in learning English pronunciation 39

2.2 Results from the interview 40

DISCUSSION 44

PART 3: CONCLUSION 45

1 Summary of the study 45

2 Limitations of the study 45

3 Recommendations 46

3.1 Recommendations for students 46

3.3 Recommendations for researchers 46

REFERENCES 48

APPENDIX A 50

APPENDIX B 54

APPENDIX C 56

APPENDIX D 58

APPENDIX E 64

PART 1: INTRODUCTION

I RATIONALE OF THE STUDY

Nowadays, in the increasing development of globalization and industrialization, English has become the second language or even the official language in numerous countries English is also a primary language of instruction in vital equipment such as computer, mobile phones and the Internet Under such circumstances, mastering English is not only an advantage but also a requirement, especially for young labor force all over the world In order to meet the requirement, they need to devote a great amount of time and effort to learn this language

Inevitably, learners have difficulties learning different aspects of English In terms of vocabulary, it is common that learners gradually lose their interest and get confused as they try really hard to remember new words and their pronunciation but eventually forget and then fail to use them when they need to The traditional method

of learning pronunciation like writing down phonetic transcription or repeating after teachers may make student feel bored and soon forget how to pronounce words properly Therefore, a more effective way is of necessity, especially which could raise student’s interest This demand has led to the development of various software in In the context of Vietnam, English Language Speech Assistant (ELSA), a tool to improve pronunciation, is becoming a popular application

At Hanoi Pedagogical University 2, however, ELSA application is not widely used by students, even English-majored ones while they seem to have difficulties in learning English and in particular pronunciation With the aim of introducing students

in English Language Teaching an effective learning method to enhance their pronunciation in detail, and create a habit of using technology in their study in general, the author has conducted the research: “A Case Study on the Effectiveness of Elsa on Learning English Pronunciation of First Year English Language Teaching Major Students at Hanoi Pedagogical University 2”

II AIMS OF THE STUDY

Ultimately, the researcher aims to introduce ELSA to students and examine its effectiveness in their English pronunciation Suggestions for better use of the technology are also provided

III RESEARCH QUESTIONS

To achieve the goals of the study, the researcher focuses on finding answers to these following questions

Question 1: What are the students’ attitudes towards the implementation of ELSA in learning English pronunciation?

Question 2: To what extent does ELSA help to improve students’ pronunciation?

IV THE SCOPE OF THE STUDY

The study is on the use of ELSA application to learn pronunciation The population of the study is limited to 10 students of K43A-English Language Teaching Major at Hanoi Pedagogical University 2

V METHODS OF THE STUDY

In this study, the researcher has decided to use the case study approach, along with the following data collecting instruments: Survey questionnaire, classroom observation and interview

VI SIGNIFICANCE OF THE STUDY

Conducting this research, the researcher hopes that its results will be beneficial

to the following:

LEARNERS: The results will provide the students with knowledge on learning

English pronunciation skills and also take a different look at learning with technology

in general and learning with smartphone’s applications to be specific The results of the study can be considered as an input for readers who want to expand their knowledge about English pronunciation, with the help of technology, learning can be

so much more interesting than before

TEACHERS: The given data of this study will be a reference for not only teachers

who are interested in applying technology in teaching English pronunciation skills but also teachers who are finding a new approach to vary their teaching techniques

RESEARCHERS: The results of this research will be beneficial to researchers who

are fond of Speech Recognition in general and the effectiveness of Speech Recognition in learning English pronunciation in particular

VII DESIGN OF THE STUDY

This thesis is divided into 3 main parts as follows:

PART 1: “INTRODUCTION” gives an overview of rationale, objectives of the

study, research questions and the scope of the study, the significance of the study and the design of the study

PART 2: “DEVELOPMENT” consists of three main chapters:

Chapter 1: “Literature Review” provides the background knowledge of

pronunciation, the background knowledge of speech recognition technology and using

of speech recognition technology in learning English pronunciation Some related researches conducted by former scholars and professionals are also illustrated

Chapter 2: “Methodology” describes in detail the participants, research

instruments employed and the procedure of collecting and analyzing the data

Chapter 3: “Findings and discussion” analyzes gathered data by research

instrument and further discussion of the data

PART 3: CONCLUSION includes three sections:

Section 1: Summary of the study reviews the research outcomes and presents

a brief description of the study

Section 2: Limitations of the study reveals in detail the difficulties in process

of conducting this study

Section 3: Recommendations gives suggestions for further study and the researchers who want to conduct studies about ELSA

PART 2: DEVELOPMENT CHAPTER I: LITERATURE REVIEW

A Pronunciation

1 Definition of pronunciation

Pronunciation plays an important role in improving our skills in English It becomes one of elements for improving communication skill Without having enough knowledge about pronunciation, communication will end up in unpleasant situations and make students feel difficult to convey their ideas There are numerous ways to define pronunciation

In Cambridge Academic Content Dictionary, pronunciation means the way in which a word or letter is said, or said correctly, or the way in which a language is spoken

According to Richards, J C., & Schmidt, R (2002, p 429) pronunciation is the way a certain sound or sounds are produced and pronunciation stresses more the way sounds are perceived by the hearer

(Harmer, 2007) states that Pronunciation is the way we make sounds of the language how and where we place the stress, and how we use pitch and intonation to show how we are feeling and what we mean”

The following diagram (Kelly, 2000) illustrates the main features of pronunciation

Figure 1 Features of pronunciation (Kelly, How to teach Pronunciation, 2000)

As can be seen from the diagram above, pronunciation consists of phonemes and suprasegmental features, the phonemes include two main parts are consonants and vowels while suprasegmental features deals with intonation and stress Besides, pronunciation is a really broad subjects with various aspects, in the scope of this study, the researcher mainly focuses on two features which are vowels and consonants

1.1 English vowel sounds

Vowels, according to (Roach), “The most common view is that vowels are sounds in which there is no obstruction to the flow of air as it passes from the larynx

to the lips”

Celce-Murcia, M., Brinton, D., and Goodwn, J (1996) defines vowels as

"sounds in which there is continual vibration of the vocal cords and the air stream is allowed to escape from the mouth in an obstructed manner, without any interruption.'"

English vowels consists of : Monophthongs/ pure vowels : Single sounds (Long vowels and short vowels), Diphthongs: Two vowels in one syllable and Triphthongs: three vowels pronounced in a single syllable

1.1.1 Monophthongs

English monothongs includes short vowels (æ, e, ʊ, ɒ, ʌ, ɪ, ə) and long vowels (iː, ɜː, ɑː, ɔː, uː), there are 12 different sounds in total and all the monothongs can be classified in terms of:

+ The length of vowels: Long vowels are longer than short vowels in similar

contexts The symbols consist of one single vowel plus a length mark made of two dots They are different from short vowels not only in length but also in quality, resulting from differences in tongue shapes and lip positions

+ Tongue position and shape (Front – Central – Back): Refers to the horizontal

position of the tongue in regard to the front or back of the mouth

+ Tongue height: High (Close) – Mid – Low (Open): In concern with the vertical

position of the tongue towards palatal area (low or high)

+ Lips position/ Degree of lip rounding: Refers to position of lips during the

progress of produce the vowels There are two lips position: Rounded and unrounded

Figure 2 English monothongs (Based on the Diagram of Cardinal Vowels (Jones,

1962))

1.1.2 Diphthongs

A diphthong is a glide from one vowel to another vowel, and the whole glide acts like one of the long simple vowels In terms of length, diphthongs are similar to long vowels The most vital thing to remember about all the diphthongs is that the first part is much longer and stronger than the second part Therefore, the second part is shorter and quieter There are 8 dipthongs in total: ɪə, eə, ʊə, eɪ, aɪ, ɔɪ, əʊ, aʊ

Diphthongs can be classified into two types: Centering (ending in /ə/) and Closing (ending in /ɪ/ and ending in /ʊ/)

1.1.3 Triphthongs

A triphthong is a glide from one vowel to another vowel and then to the third, all of these vowels are produced rapidly without interruption There are five triphthongs in English They are composed of the five closing diphthongs with /ə/ added at the end: eɪə, aɪə, ɔɪə, əuə, auə

1.2 English consonant sounds

Kell (2000, pp 47-53) states that “Consonant is formed by interrupting, restricting or diverting the airflow in a variety of ways”

Consonants are classified according to:

+ Voicing

+ The manner of articulation

+ The place of articulation

+ The force of articulation

Table 1 English Consonant Phoneme

1.2.1 English consonants classification

1.2.1.1 Voicing

+ Voiced consonants: The consonants which are produced when the vocal cords are

vibrating

+ Voiceless consonants: The consonants which are produced when the vocal cords

are not vibrating

1.2.1.2 Manner of articulation

Refers to the nature of the obstruction

+ Stop (Plosive): A total closure of the articulators involved in order that the

airstream is not able to escape through the mouth: / p, t, k, b, d, g /

+ Fricative: The close approximation of two articulators in order that the airstream is

partially obstructed and turbulent airflow is produced: / f, v, Ɵ, ð, h, z, s/

+ Affricate: The combination of a plosive sound and a fricative sound: /tʃ/ /dʒ/

+ Nasal: Similar to plosives, except that there is a full closure in the mouth, but as the

velum is lowered the air can escape through the nasal cavity [m], [n], [Ƞ]

+ Approximant: An articulation in which one articulator is close to another, but

without the vocal tract being narrowed to such an extent that a turbulent airstream is produced /w,r, j/

Lateral: Obstruction of the airstream at a point along the center of the oral tract, with

incomplete closure between one or both sides of the tongue and the roof of the mouth /l/

1.2.1.3 Places of articulation

+ Bilabial: Made with the two lips (/p/, /b/, /w/, /m/)

+ Labiodental: The lower lip articulates with the upper teeth (/f/, /v/)

+ Dental: The tongue tip articulates with the upper teeth (/Ɵ/, /ð/)

+ Alveolar: The tongue tip articulates with the alveolar ridge (/t, d, n, l, s, z/)

+ Palato-Alveolar: The tongue tip and the back of alveolar ridge (/tʃ/ /dʒ/, /r/)

+ Palatal: The front of the tongue and the hard palate (/j/)

+ Velar: The back of the tongue and soft palate (/k, g, Ƞ/)

+ Glottal: The sounds produced when the air passes through the glottis as it is

narrowed: [h]

B Speech recognition technology and ELSA

1 Definition of speech recognition

ASR (Automatic speech recognition) falls within the area of Computer Assisted Language Learning (CALL) (Levy, 1997) CALL system has become very popular in recent years due to its advance in speech recognition that allow students to get understood better by computer, CALL offers student extra learning time and a stress-free learning environment, which mainly because CALL is often used in selft – study methods

ASR is a cutting edge technology that allows a computer or even a hand-held PDA (Myers, 2000) to identify words that are read aloud or spoken into any sound-recording device ASR technology has been widely used for business dictation or special needs accessibility, its market presence for language learning has increased dramatically in recent years (Aist, 1999)

With the integration of ASR technology, these systems, which can be referred

to as CAPT (Computer Assisted Pronunciation Training) systems The computer

understands the student’s speech and reacts spontaneously, thus making the learning process more realistic and engaging, the CAPT systems can also give feedback on the quality of the student’s speech Early CAPT systems used the template-based recognition systems which perform pattern matching and most of these programs do not give any feedback, these systems can only give a simple comparison between students’ speech sound waves and the templated ones Later then, ASR programs adopted the Hidden Markov Model (HMM) algorithms That is, a speech recognizer based on HMM computes how close the phonemes of a spoken input are to a corresponding model, based on probability theory High likelihood represents good pronunciation; low likelihood represents poor pronunciation (Larocca, S T., Moagan

J J., & Bellinger S M, 1991, pp 295-310)

The most recent methods that today ASR software use is Artificial Intelligence (AI) which is the latest trend in the Industrial Revolution 4.0, by using AI, the programs can learn every time they are used, the AI systems are programmed to adapt and develop by itself, this is the key solution to the individualism of all users

2 An introduction to ELSA Speak

ELSA stands for English Language Speech Assistant, ELSA is an AI-powered program that allow users to practice and improve their pronunciation by a set of lessons and exercises that are evaluated on ELSA’s server ELSA was first released for testing in 2015 and then in 2016, ELSA was debuted during SXSW (a component

of the South by Southwest® (SXSW®) family of conferences and festivals)., the program won the SXSWEdu launch competition Until now, ELSA parent company has released two version of the application, which are on Google Android (A mobile phone operating system by Google LLC) and Apple IOS (A mobile phone operating system by Apple Inc.)

In 2017, ELSA reached 2 million users from over 100 countries, more than 100 million practices, 81% of ELSA users are reported to feel more confident after they start using ELSA (ELSA Speak Blog, 2017), ELSA corp was also mentioned by Forbes magazine in the list of 4 companies that are using AI to transform the world ELSA is free to download from both Google Play (The official application store for the Android operating system) and App Store (The official application store for the IOS operating system), ELSA offers users four different features which are:

+ Skills: This is the feature that contains all of the lessons about pronunciation theory,

ELSA guides users through 12 lessons, each lesson is about different skills, the difficulty increase gradually as users go from the first to the last lesson

+ Topics: In this sections, users can choose the topic that they want from 21 different

topics for different purposes

+ Dictionary: This is the most popular feature of ELSA, users can type any word,

phrase or sentences and ELSA will give users phonetic transcription, the audio of correct pronunciation and then ELSA will help users test their pronunciation, ELSA can point out specifically the user’s mistakes and after that, all of the texts are filled in with different colors, green stands for correct, yellow is for quite correct and red is for incorrect

+ Progress: In this area, ELSA display the user’s proficiency level and ELSA also

offers a Proficiency level/ Assessment Test with a detailed report for each pronunciation skills

The detailed report gives information about: Schwa sound, the TH sounds, ending sounds, the short and long /i:/, the /r/ sound, consonant cluster, the /l/ and /r/ confusion, the /ʃ/, /ʒ/, /tʃ/ and /dʒ/ sounds, the /s/ and /z/ contrast, the /p/, /t/ and /k/ sounds, /w/ and /v/ and /b/

3 The use of speech recognition in learning pronunciation

ASR off ers many advantages for language learners, this technology gives teachers a chance to find out individual problems of the learners so that teachers will

be able to know which are the student strong points or weak points so that teachers can understand their students better and modify their curriculum to be suitable for each level of students ASR can also help students who are afraid from practicing speaking in public to improve their speaking skill (Neri, A., Cucchiarini, C & Strik H.(, 2001, pp 40–48)

Most of the studies which have investigated the effectiveness of ASR programs have showed that this technology can be effective For instance, Neri (2008) investigated if a CAPT system can help learners improve word-level English pronunciation skills at a level comparable to that achieved through traditional teacher-led training Their results showed that the pronunciation quality of isolated words improved significantly after using ASR-based method (Kim) examined the reliability

of ASR software which was used to teach English pronunciation The oral production

of 36 students was compared to pronunciation scores determined by native English

speaking instructors Although the results indicated that ASR technology is still not as

accurate as human analysis, the author concluded that the software may be useful for

student practice with certain aspects of pronunciation (Denis Liakin, Walcir Cardoso,

Natallia Liakina, 2013)

In short, ASR-based method or ASR software cannot be 100 % correct yet, the

systems have not been able to identify a number of aspects such as intonation, accent

or emotions The average score for ELSA is reported to be around 90%, ELSA user

are reported to increase their pronunciation score by 40% within less than 4 weeks

CHAPTER II: METHODOLOGY

This chapter presents detailed description about research method, research procedure, data collecting instruments and data analysis description

A case study is a unique way of observing any natural phenomenon which exists in a set of data (Yin, 1984) Usually in a case study, only a small number of subjects of interest are examined in detail

1.1.1 Rationales behind the case study

The case study approach was chosen due to the following reasons:

Firstly, case study is considered to be the most comfortable and least stressful method for the novice researcher, the researcher was able to conduct a case study with

a small number of participants and focus on them

Secondly, case study is a qualitative method that makes it possible for the researcher to have an in-depth look of the issue and it may offer some insight on how

to conduct further research in the area

Finally, because of the second teacher training month (as the researcher is a senior English Language Teaching student), the long distance and the tasks of trainee teacher made it really difficult for the researcher to be at HPU2 frequently, therefore, case study is the most suitable method for both the researcher and participants as well

1.2 Research procedure

There are many different models for a case study to follow and in this study, the researcher chose the steps suggested by (Chorev, 2013) from University of

Pennsylvania – The Wharton School, Sydney Australia According to Raz Chorev, 7 steps are conducted in the model of case study

1.2.1 Describing background

In the first step, the researcher need to provide a brief background describing the situation, the people or the organization involved Not too specific, but a description good enough for anyone to understand fully the context of the case study

10 students of K43A – English Language Teaching at HPU2 were chosen to be the participants of this study, they all had different problems in learning pronunciation and want to improve their skills, especially, they were very excited to use ELSA because they heard about it a lot and this is also the first time they use smartphone AI technology to support their learning

1.2.2 Identifying problems

In this section, the researcher identifies and clearly articulate the problem that the study is going to solve The challenge the researcher facing, and would like to solve Quantify the situation and provide specific description of the issues The aim of this section is to get the reader to buy-in and empathize, and create an emotional connection

The focus group of this study appeared to have lots of difficulty in pronouncing English words correctly and fluently, they were always willing to learn and do have great passion in English At high school, the curriculum that they followed does not really focus much on speaking in general and pronunciation in particular, thus, students might have excellent reading, listening or writing skills but speaking skills were still a challenge for them And the very first start of speaking well is mastering pronunciation which is also what this research is about

1.2.3 Vision & objective(s)

In this section, the researcher shows what exactly does the study want to achieve, and how the researcher sees the situation, and what is an ideal outcome to expect

The main objectives of this study are to help the participants enhance their English pronunciation skills and also examine if their speaking skills could be

to introduce a new method of learning pronunciation which involves technology as applying technology in education is currently a trend all over the world

Table 2 The schedule of implementing ELSA in 6 weeks

Introduction to ELSA Pre-ELSA Assessment Test

2 23/03/2018 How to use basic features of ELSA

2 1 27/03/2018 Introduction to the implementation of ELSA

Lesson 2 : /ʊ/ and /u:/

Lesson 3 : /e/ and /æ/

2 06/04/2018 Lesson 4 : /ɪ/ and /i:/

Lesson 5 : /ɒ/ and /ɔː/

Lesson 6 : /ə/ and /ɜː/

Lesson 7 : /eɪ/, /aɪ/ and /ɔɪ/

Lesson 8 : /əʊ/ and /aʊ/

2 12/06/2018 Lesson 9 : /ɪə/, /ʊə/ and /eə/

1.2.5 Implementing case study

The implementation description is among the most detailed steps This section includes various factors, such as resources, technology, people and processes involved

in the implementation

Following the researcher’s instruction, the participants spent about 1 hour in the evening to learn English pronunciation skills with ELSA in every Tuesday and Friday during the period of 6 weeks At first, the participants were a little bit confused with the usage of ELSA as well as how to study effectively all of the information given in the guiding book, in every session that the participants took, the researcher always stayed available for them to contact and help them solve any problems occurred

For every two weeks, a Q & A session was conducted, the researcher read about the problems that participants faced when using ELSA and the author also gave answers

to the questions about how to learn English pronunciation better and how to become a better speaker of English, etc After every session, there was a practicing sheet which

score were in between 70 to 75/100 Having used ELSA for a long time, the researcher found out that these numbers were the suitable scores for practicing with ELSA, only when a little pressure of ranking is put on, the participants would pay more attention and do their best to achieve that score After each session, the researcher asked the participants to take screen captures of their practicing in ELSA as

a proof so the author would be able to know that if the students were doing their tasks

or not The researcher also gave feedbacks for the students after all the on-campus classes that the researcher played the role of an observer so that they could improve their speaking skills week by week

1.2.6 Results and evaluation

The results of this study were collected from all the following sources: The pre and post ELSA Assessment Test, the pre and post survey questionnaire, the SOLOM Assessment score and the interview Based on the results, the researcher concluded that the implementation of ELSA in learning English pronunciation had effectiveness

on all of the participants’ pronunciation skills, some participants had significant improvements in their pronunciation skills and some participants stated that they felt much more confident when it comes to pronunciation thanks to 6 weeks of using

ELSA

2 Data collection instruments

Questionnaires, classroom observation and interviews are considered useful and feasible implements for the researchers to find out the answers for the research questions Using survey questionnaires, classroom observation and interviews give the researcher useful, qualitative and dependable results

2.1 Survey questionnaire

2.1.1 Reasons for using questionnaire

Using questionnaires is one the most common instrument in collecting data for

a research because a big amount of information could be collected from a large number of people in a short period of time and in a fairly cost effective way Besides, the results of the questionnaires can be easily quantified and quickly analyzed and by either a researcher or through the use of a software package

2.1.2 Description of survey questionnaire

 The pre - survey questionnaire (Appendix A)

The pre - survey questionnaire used in the thesis consists of 14 questions which are divided into 2 parts

Part I includes 7 question used to collect personal information of the students

who were given the questionnaires

For the students who are not interested in joining in the research, the researcher asked them to fill in only their genders and the number of years that they have been learning English

For the student who are interested in joining in the research, the researcher asked them to fill in enough information for the researcher to contact

All of the personal data collected was used for contact purpose only

Part II includes the next 5 questions about the participants’ opinions on

learning English pronunciation skills, which method do they use or find out whether they were having difficulty in learning English pronunciation or not

In this section, question number 12 is a Yes-No question used to investigate whether the participants have problem in learning English pronunciation or not

For those who choose the answer “Yes”, they are asked to choose the answer from the list of problem in learning English pronunciation pointed out by the researcher

For those who choose the answer “No”, they are asked to continue answering the next questions

Part III includes the last 2 questions which are aimed to find out the

participants’ experience on using ELSA in learning English pronunciation skills

In this section, question number 14 is a Yes-No question used to investigate whether the participants have used ELSA in learning English pronunciation skills

For those who choose the answer “Yes”, they are asked to continue answering the next questions about their experience using ELSA in learning English pronunciation skills

For those who have not used ELSA in learning English pronunciation skills, they are asked to answer some questions to reveal their experience on the speech recognition technology and application’s instant feedback on users’ pronunciation

The post - survey questionnaire used in the thesis include 10 questions which are divided into 2 parts

Part I includes questions 1-10 aiming to find out participants’ attitudes towards using

ELSA to learn English pronunciation, in this section, the researcher wants to know that if the participants think ELSA is effective and useful or not

Part II includes questions 11 – 14 which are used to identify the problems

participants facing when using ELSA to learn English pronunciation Along with multiple choice questions, question number 10 offers an own answer from the

participants

2.2 Classroom observation

2.2.1 Reasons for using classroom observation

Using classroom observation could help the researcher have a practical, authentic look on the improvement of the students week by week, as the study focused mainly on the self-study of the participants, therefore, it was very important for the researcher to know if the participants are making any progress or not After every session, the researcher made a weekly report about the changes and improvements of the participants

2.2.2 Procedure of classroom observation

Once a week, the researcher attended the speaking class that all of the participants are taking as an observer During the 6 week period, the researcher noted down common pronunciation mistakes which the participants might make and then compared the results of each week to see if the number of mistakes decrease as amount of time the participants using ELSA increases

As an observer in class, the researcher wrote down all the of the pronunciation

mistakes that most of the participants made, the weekly report (Appendix C) was

adapted from Student Oral Language Observation Matrix (SOLOM) which is used to rate the quality of student’s speaking skills SOLOM was originally developed by the San Jose Area Bilingual Consortium and had undergone revisions with leadership from the Bilingual Education Office of the California Department of Education

Table 3 The SOLOM Assessment Score

2.3.1 Reasons for using interview

Using interviews is considered a qualitative research tool that enables the researcher to learn about things that cannot be directly observed about personal feelings, perceptions and opinions because interviewees are not influenced by others

in group At the same time, the interviews are interactive, it allows researchers to suggest data if the original answer is vague, incomplete or non-specific The researcher could clarify the terms if they are not clear to people who are interviewed

to obtain information Participants, who have qualifications and enthusiasms, will be ready to provide the actual documents for research

For a detailed and logical schedule, a guiding book (Appendix D) for 6 weeks

was given to each participant The guiding book comprises 20 lessons in total which are divided into 8 sessions, there are 2 sessions per week, the first session of each week contain 3 lessons and the second session consists of 2 lessons There were practicing sheets for every week, the sessions usually take place on Tuesday and

3.1 Participants

 The researcher - Observer

The researcher took the responsibility of a supervisor and a supporter in student’s implementing of ELSA in learning English pronunciation

 Students

10 first year students of English Language Teaching major at Hanoi Pedagogical University 2 were chosen to be the participants of the study The vast majority of these students had learnt English at least 10 years They are estimated to

be at intermediate level in English proficiency The reason why the researchers chose these students as the contributors of the study was that they inexperienced students They had applied various methods and did not have much experience in learning English pronunciation

4 Data analysis description

4.1 The ELSA Assessment Test

The participants were asked to take the ELSA Assessment Test at the first and the last session, ELSA gave instant feedback and a detailed report for every user, the audio of users’ speech was also automatically recorded The researcher collected all of the participants’ score and detailed reports, the audio of the participants speaking was also recorded again for evaluation based on the SOLOM Assessment with the consults

of the supervisor and HPU2’s foreigner teachers

4.2 Classroom observation

As an observer, the researcher made weekly reports of common pronunciation mistakes that the participants made, the researcher also assessed the students’ English proficiency based on the SOLOM with a special attention to pronunciation The score from the SOLOM was used to compared with the amount of time that the participants had been using ELSA Based on the results, the researchers could conclude that the students’ pronunciation skills were enhanced weekly

4.3 Survey questionnaire

The data collected from the pre and the post questionnaire was carefully analyzed with the help of Microsoft Excel, there were two type of answers, the first type of answers comes with two options: Yes/ No and the second type gives the students’ various choices and also their own answers if it was a complex question

With the application of Microsoft Excel in analyzing data, the researcher was able to

get the survey questionnaire’s results quickly and precisely

4.3.1 The pre-survey questionnaire

The pre-survey questionnaires were given to 30 first year English Language

Teaching majored students at HPU2 in 14/03/2018, there were 3 male students and 27

female students They were from 2 classes: K43A – English Language Teaching and

K43B – English Language Teaching, most of them had been learning English for at

least 7 years The pre-survey questionnaire focused with these themes:

 Students’ attitudes towards English pronunciation skills

 Students’ difficulties in learning English pronunciation skills

 Students’ experience with ELSA

4.3.2 The post-survey questionnaire

After 6 weeks of implementing ELSA, the researcher gave the post-survey

questionnaire to 10 participants of the study in 02/04/2018, the post-survey

questionnaire focused on the students’ attitudes towards the implementation of ELSA,

the benefits of ELSA and the flaws of the application in learning English

pronunciation

To present the data of the pre and post questionnaire in a comprehensive way,

the researcher chose to use Microsoft Excel’s calculator and graph illustrator

4.4 Interview

The post-interview was conducted in 02/04/2018 with the aims of collecting

more information, more feedbacks from the participants, the researcher could be able

to have a practical perspective and the results would be more clarified The data was

classified into two types : Yes/ Maybe or No and students’ own answers

CHAPTER 3: FINDINGS AND DISCUSSION

This chapter analyzes the data gathered by research instrument and further discussion of the data It is divided into two main parts: Findings part and Discussion part

FINDINGS

The findings of this study are divided into two sections The first section provides the progress of the participants through 6 weeks (the pre and the post ELSA Assessment TEST, the weekly classroom observation) The second section reveals the results in detail about the students’ attitudes towards using ELSA in learning English pronunciation skills as well as the challenges that the participants faced during the implementation of ELSA in learning English pronunciation (the pre and post-survey questionnaire and the interview)

1 The results of using ELSA in learning English pronunciation skills in 6 weeks 1.1 Description

In this section, the researcher describes the implementation of ELSA as supporting tool for learning English pronunciation skills during 6 weeks In the process of conducting this study, the researcher played the role of an observer and a

supporter of students’ improvement progress A Facebook group was created to make

it easier for both the researcher and the participants to communicate and to conduct an

Q & A session for every two weeks

( Link to the Facebook group : https://www.facebook.com/groups/358651551289520 )

1.1.1 Week 1 (19/03/2018 – 25/03/2018)

When the researcher first introduced ELSA to the students, they were very excited because most of them had heard about it at least once At first, the students were a little bit confused about how to download the official ELSA application and what are the features that the study would focus in To help the participants get more familiar with the application, the researcher made a brief introduction of ELSA and also detailed instruction on using ELSA basic features

In the first week, the participants were asked to take the ELSA Assessment Test to check their English Proficiency, the researcher also took their recordings to re-

evaluate with the help of the supervisor and HPU2’s foreigner teachers based on SOLOM Pronunciation Assessment The results are showed in the Table 4 below

Table 4 The comparison of ELSA score and SOLOM score from the participants

It was such a relief for the researcher because he thought would not be able to assess all of the students’ performances due to the lack of time in a speaking class or being afraid that some students would not feel comfortable to talk in front of class The results from the observation was based on the SOLOM Assessment The results are showed in the Table 5 below

 Week 1 observation: SOLOM Assessment results

Table 5 The SOLOM Assessment results of Week 1 (Scale from 0 to 25)

specific curriculum for the next 4 weeks, with 20 lessons about all of the vowels and consonants in English, the participants were eager to start learning As freshmen in university, the participants had a lot of questions about learning English in general and learning pronunciation in particular, the researcher spent one night reading all of questions that the participants sent and then answered all of them one by one, the students said that they really felt more confident to learn English pronunciation

 Week 2 observation: SOLOM Assessment results

Table 6 The SOLOM Assessment results of Week 2 (Scale from 0 to 25)

In the third week classroom observation, the researcher noticed some minor changes in the participants’ pronunciation, many of them paid more attention to the length of vowels and how to move their tongue to the right position Even though,

there were many pronunciation mistakes made by the participants

 Week 3 observation: SOLOM Assessment results

Table 7 The SOLOM Assessment results of Week 3 (Scale from 0 to 25)

any confusion about usage of the guiding book In the Q & A session, the participants mainly wrote about the limitation of ELSA, especially the microphone, they reported that different devices gave different results for the same person speaking the exact sentence

Week 4 observation: SOLOM Assessment results

Table 8 The SOLOM Assessment results of Week 4 (Scale from 0 to 25)

Week 5 observation: SOLOM Assessment results

Table 9 The SOLOM Assessment results of Week 5 (Scale from 0 to 25)

Test, the audios were also recorded for re-evaluation using SOLOM Assessment The results are showed in Table 10 below

 The results of Post-ELSA Assessment Test in comparison with SOLOM score (ELSA: Scale from 0 to 100, SOLOM: Scale from 0 to 5) (P: Participant)

Table 10 The comparison of ELSA score and SOLOM score from the participants

Week 6 observation: SOLOM Assessment results

Table 11 The SOLOM Assessment results of Week 6 (Scale from 0 to 25)