an investigation into the impact of speechace in developing pronunciation competency for non english majored students at hoa sen university

VIETNAM NATIONAL UNIVERSITY - HOCHIMINH CITY UNIVERSITY OF SOCIAL SCIENCES & HUMANITIES FACULTY OF ENGLISH LINGUISTICS & LITERATURE AN INVESTIGATION INTO THE IMPACT OF SPEECHACE IN DEVEL

INTRODUCTION

B ACKGROUND TO THE STUDY

Pronunciation is an important language component that one needs to master if he/she wants to communicate effectively Unfortunately, teaching pronunciation in Vietnam has not been taken seriously since Vietnam’s educational system favors receptive skills over productive ones (Vu, 2016) In middle schools and high schools in Vietnam, teaching and learning English emphasizes on grammar, reading, and listening with lots of exercises while neglecting speaking and pronunciation These two language components are also excluded from the tests Moreover, pronunciation is taught very briefly in the audiolingual method, which cannot motivate students As a result, after seven years of learning English, from grade 6 to grade 12, it is very hard for students to carry out even the simplest everyday conversation If pronunciation had been focused on, students would have found joy in learning to speak English and they would have wanted to speak more Fortunately, in the era of computer-assisted language learning (CALL), there are potential approaches in teaching and learning pronunciation thanks to the emergence of automatic speech recognition (ASR) Since its existence, ASR has been widely applied in everyday life such as virtual assistants (Siri, Google Assistant, Cortana) in smart devices that listen and execute people’s commands Later, ASR has also been applied in the education sector, especially in pronunciation training A notable example of ASR in training pronunciation is the software Praat which analyses human speech in detail phonograph to show the rising and falling of the graph the person is speaking to see if it matched with the native speaker’s utterance Other learning English software which integrates ASR in their features such as Rosetta Stone, Duolingo are also popular amongst English learners

Recently, there is the presence of SpeechAce, an ASR program which compares users’ speech to the native speaker’s ones Such ASR software like Praat and SpeechAce have paved new paths in seeking a better approach to teaching and learning pronunciation Thus, further studies need to be done on software similar to those to introduce new methods in teaching pronunciation to students and make their learning of pronunciation more motivating and meaningful.

A IMS OF THE STUDY

Despite English's dominance in global communication, non-English majors at Hoa Sen University face challenges in acquiring accurate pronunciation due to limited exposure and practice While the English File textbook employed in the classroom includes some pronunciation instruction, it is insufficient Formal pronunciation teaching is time-consuming, leading to reactive instruction in response to student errors or occasional neglect when intelligibility remains unaffected However, this approach has proven ineffective, as students continue to repeat pronunciation mistakes despite corrections.

To tackle the previously stated problems, there calls for an innovative method in teaching and learning pronunciation and SpeechAce may offer such method SpeechAce, as an innovative language learning tool, utilizes speech recognition technology to provide immediate feedback on pronunciation, which could be particularly beneficial for learners without regular access to native speakers or specialized language instructors As a result, the study titled “an investigation into the impact of SpeechAce in developing pronunciation competency for non-

English-majored students at Hoa Sen University” aims to answer the following questions:

1 To what extent does SpeechAce improve students’ pronunciation competency?

2 What are the students’ perceptions toward learning pronunciation with SpeechAce?

S IGNIFICANCE OF THE STUDY

The findings of this study could have significant implications for language learning strategies at Hoa Sen University and similar institutions If found effective, SpeechAce could be adopted more widely, leading to improved pronunciation skills among students, thereby enhancing their overall English competency Additionally, the results could inform educators and curriculum designers about the potential benefits of incorporating specialized language learning technologies in their teaching method Additionally, the results could inform educators and curriculum designers about the potential benefits of incorporating specialized language learning technologies in their teaching methods

Beyond the immediate context, this research contributes to the broader field of educational technology and language acquisition By examining the role of technology in developing pronunciation skills, the study adds to the understanding of how digital tools can aid in overcoming language learning challenges, thus supporting the goal of fostering multilingual competencies in an increasingly interconnected world.

S COPE OF THE STUDY

The study is done with two general English classes with a total of 74 participants from Hoa Sen University Although the study is on pronunciation, it does not cover all aspects of pronunciation Rather, this study is limited to using SpeechAce for students to practice stress, identification of a number of syllables, vowel, consonants sounds, phonemic transcript since these are the pronunciation aspects that Vietnamese English learners usually have trouble with Moreover, these are the pronunciation features which are included in the textbook English file which both classes are using As a result, testing these pronunciation features only would guarantee the reliability and validity of the study, which is done in one semester of approximately three months.

O RGANIZATION OF THESIS CHAPTERS

The present thesis is divided into five sections Following the introduction is section II, literature review, which defines related terminologies and aligns articles that the author based on to develop the study Section III describes the methodology, which gives detailed information on how the study is carried out, the participants, and the research site Section IV presents the findings and interprets them Section V concludes the interpretation from the findings, suggests implications, and acknowledges limitations of the study.

LITERATURE REVIEW

C OMPETENCY IN E NGLISH PRONUNCIATION

2.1.1 Definition of competency in English pronunciation

English pronunciation competency, as defined in the realm of language learning, entails the ability to accurately produce the sounds of English, including its vowels, consonants, and stress patterns, in a way that is intelligible to listeners (Celce-Murcia, Brinton, & Goodwin, 2010) It encompasses not only the precise articulation of individual phonemes but also the mastery of the rhythm, stress, and intonation patterns that are characteristic of fluent English speech (Jenkins, 2000; Gilbert, 2014) This competency extends beyond mere phonemic accuracy; it also involves understanding and applying the nuances of connected speech, such as linking and elision, which are pivotal for natural and comprehensible spoken English (Morley, 1991) Furthermore, Derwing and Munro (2005) emphasize the importance of intelligibility in pronunciation, highlighting that the primary goal is effective communication rather than the imitation of native-speaker norms Therefore, English pronunciation competency is a multifaceted skill set crucial for successful communication in English, focusing on intelligibility, comprehensibility and interpretability rather than adherence to native-like pronunciation The revised version of the CEFR in 2017 further confirmed such assertion, in which the term phonological control is introduced and focused on crucial pronunciation feature such as intonation, rhythm, stress, control of sounds to maintain intelligibility While comprehensibility and interpretability take into consideration features such as context, lexical, grammatical aspects of L2 speech, and social knowledge of speakers and hearers (Pennington & Rogerson-Revell, 2019), intelligibility is closely related to pronunciation There are various definitions of intelligibility Smith and Nelson (1985) defined it as the interaction between speakers and listeners As for Munro et al (2006) intelligibility is “the extent to which a speaker’s utterance is actually understood” (p.112) Being intelligible is also considered in scoring the pronunciation criterion in the IELTS speaking test Specifically, according to the IELTS Speaking Band Descriptors, for a candidate to get band 8 in pronunciation, he must be “easy to understand throughout and his L1 accent has minimal effect on intelligibility” On the other hand, he will get band 2 if his “speech is often unintelligible” (ietls.org) Harmer (2016) also stressed the importance of being intelligible to be understood by other speakers of English, regardless of your accents If being intelligible is crucial to be competent in English pronunciation, then what should be the main focus in teaching pronunciation to L2 learners? Jenkins (2000) mentioned the term “Lingua Franca Core”, which includes crucial phonological features such as certain consonants, vowels, stresses in English pronunciation that English learners need to master in order to maintain mutual intelligibility amongst speakers and hearers

While native-speaker pronunciation is not the ultimate goal for English learners, they must master crucial pronunciation features Teaching pronunciation should prioritize prominent components, making lessons more accessible for both educators and students By focusing on these essential elements, pronunciation instruction can become more manageable and effective.

2.1.2 Vietnamese English learners’ difficulties in pronunciation

Despite sharing the Latin Alphabet, English and Vietnamese possess distinct phonological features that present pronunciation challenges for learners of both languages Vietnamese, with its 29 letters, employs a consistent pronunciation system where each letter represents a specific sound For instance, "a" is pronounced as /ɑː/, "c" as /k/, "b" as /b/, and "d" as /j/ By mastering these letter sounds, Vietnamese learners can effectively pronounce any Vietnamese word.

“da” /jɑː/ Such phonetical consistency is not the case in English Knowing how to pronounce the 26 letters in the English Alphabet does not promise learners with the correct pronunciation of certain English words For instance, the vowel “a” is pronounced /eɪ/, but “banana” is pronounced /bəˈnɑːnə/, “apple” is /ˈổpl/, and

“what” is /wɒt/ Even some consonants have more than one version of pronunciation, such as “c” is pronounced /s/ as in “certain”, but /k/ in “curtain”; or

“g” is /g/ in “girl” but /ʤ/ in “gym” Moreover, some words in English which contain the same clusters of letters, yet pronounced differently, such as “thought”,

“through”, “though”, “cough”, “plough”, “tough” all of which contain “ough” yet pronounced as /θɔːt/, /θruː/, /ðəʊ/, /kɒf/, /plaʊ/, and /tʌf/ The irregularity of English pronunciation is so notorious that there is even a poem about it “The Chaos” written by Gerard Nolst Trenité (1870-1946) in 1920 (wikipedia, 2021) The inconsistency in English pronunciation and spelling, along with other English phonological features which are not present in Vietnamese, such as stresses, connected speech, intonation have made it challenging for Vietnamese English learners to acquire English pronunciation

In the researcher’s teaching experience, the most noticeable pronunciation mistake that learners make is the omission of ending sounds, which is a very important feature in English pronunciation in a way that the ending sounds, in most cases, decide the meaning of a word For instance, learners may pronounce “wife”,

Vietnamese learners of English often struggle to pronounce certain English sounds, including "why," "white," and "wine," which are typically pronounced as /wai/ due to their native language's lack of aspirated consonants Vietnamese's syllable-timed nature also poses challenges, as learners tend to speak English sentences with equal loudness, lacking emphasis Additionally, connected speech, such as the phrase "pick it up," can sound unfamiliar to learners and hinder recognition Intonation is another difficulty, as learners often speak all sentence types with a flat tone Specific English sounds like "t," "ʧ," and "ʒ" are particularly challenging to pronounce, leading to mispronunciations like "tea" as "/tri:/" instead of "/ti:/.

Learners’ difficulty in pronunciation is also rooted from their lack of phonemic symbols’ knowledge Most learners are unable to pronounce English words from their phonemic transcriptions, or their awareness of such component is vague Many teachers argue that it is possible to pronounce a word properly by only hearing and repeating This may be true only when the process is repetitive and when there is a correspondence between the sound and spelling However, such correspondence is not always the case in English As a result, introducing learners to phonemic symbols will help them develop their autonomy in learning pronunciation in a way that they can say the word without hearing it Furthermore, the phonemic symbols are wildly accessible in both e-dictionary and paper one, thus when learners make mistakes in pronunciation, it is easier for teachers to explain and pinpoint the mistakes (Harmer, 2016)

So, in order to have correct pronunciation, learners need to master all the mentioned components of pronunciation and expose to English speech as much as possible Such exposure can be extensive listening to English speech in media products such as movies, YouTube videos, songs, or the listening of teacher’s talk Those exposures can help learners get used to different English pronunciation versions However, they may also confuse learners since there are so many versions of English pronunciation and learners do not know which one to base on to practice their conversation Another drawback of such exposures is that they are mostly passive listening, which means learners may only listen without responding to what they are listening to, thus they are not able to compare their pronunciation with a standardized one In case of listening to the teacher’s talk, learners may respond to the teachers and get feedback, but this is very time-consuming The teachers do not have all the time to specifically correct all students’ pronunciation

Moreover, the shyness characteristic of Vietnamese English learners may make them reluctant to respond to the teachers.

T EACHING PRONUNCIATION IN THE CLASSROOM

2.2.1 The extent of pronunciation instruction in ESL classroom

Although teaching pronunciation has received more attention in recent years, research have shown that it does not receive as much formal instruction as other areas of language (Lee, Jang, & Plonsky, 2015) Recent surveys have shown that pronunciation content only appears as add-ons and in a non-systematic manner in most commercial English language teaching textbooks (Derwing, Diepenbroek, & Foote, How Well do General-Skills ESL Textbooks Address Pronunciation?, 2012) Foote et al.’s analysis (2013) of ESL lessons in Québec, Canada found that teaching pronunciation occurred infrequently and spontaneously without any formal instruction For teachers, they also admit that they have less time to teach pronunciation First, it is because the curriculum is devoted to other language skills Second, the teach-to-test approaches also affect teachers’ decisions on which language areas to focus on to help learners do well in the tests, since most English tests examine the four English skills, while pronunciation is counted as one criterion in speaking skill Finally, teachers’ hesitation in teaching pronunciation also lies within their lack of confidence, skills, and knowledge (Pennington & Rogerson-Revell, 2019) This might be because the teachers themselves did not receive formal pronunciation training back when they were students Teachers may claim that students may acquire pronunciation through the learning of speaking and listening However, if pronunciation is taught formally and learners are aware of the English sounds, their speaking skills will be improved significantly (Harmer, 2016)

Celce-Murcia, Brinton, and Goodwin (2010) identify two distinct approaches in pronunciation teaching: analytic-linguistic and intuitive-imitative The analytic-linguistic approach involves detailed analysis of pronunciation mechanics, including articulation, phonemic symbols, and phonological features Conversely, the intuitive-imitative approach emphasizes intuitive and spontaneous instruction by teachers Despite its potential advantages, time constraints, limited content, and confidence issues among teachers often lead to the preference for the intuitive-imitative approach in pronunciation teaching.

This is also true in the researcher’s experience In most years of his teaching, he has never taught a separate pronunciation session Most textbooks he has used only integrate pronunciation as a sub-skill in speaking The only time when pronunciation is taught is when he recognizes the students have made prevalent errors For instance, students tend to put the wrong stresses in ‘comfortable’,

‘hotel’, ‘police’; or they mispronounce certain consonant sounds such as /θ/ to /t/, /ʒ/ to /s/ In those cases, the researcher took the opportunity to give students pronunciation feedback Sometimes, he also went further and talked about relevant phonological features All the instructions occurred spontaneously and informally Unfortunately, it came to his attention that the students later often made the same mistakes again For some teachers, they may have brilliant methods to explain and demonstrate certain pronunciation features However, Pennington et al (2019) suggested that adopting an intuition-based rather than evidence-based approach has its drawbacks First of all, teachers’ knowledge may be inaccurate Second, teachers may spend too much time teaching unnecessary features Finally, learners’ needs are not taken into consideration if taught in such an intuitive manner

2.2.3 Teaching and learning English pronunciation in Vietnam context

English instruction in Vietnamese state schools predominantly emphasizes theoretical knowledge, resulting in programs that prioritize receptive skills The 7-year English program lacks dedicated pronunciation sections, relegating it to self-study Formal pronunciation instruction is hindered by the dominance of other language skills in class time Despite the National Foreign Languages 2020 Project aimed at developing all language skills equally, it has been unsuccessful The English pronunciation of Vietnamese students remains difficult to understand for both native and Vietnamese English speakers This highlights the need for a renewed approach to pronunciation teaching to enhance student communication abilities.

Since teaching pronunciation has not received serious attention English teachers in Vietnam are also not well-prepared for their pronunciation teaching skills A recent study by Nguyen and Newton (2020) showed that most teachers taught pronunciation in unplanned and reactive ways This was because they were not offered formal pronunciation classes during their training As a result, students’ errors were corrected with certain repetitive techniques such as recasts or prompts However, these types of feedback do not have a long-term effect on students’ pronunciation competence If overused, they might demotivate the students’ pronunciation learning They may also prevent students from developing their autonomy Thus, the students may never be able to learn pronunciation on their own The findings from Nguyen et al.’s study (2020) also showed that the teachers expressed a strong need for more formal pronunciation training sessions which can equip them with the necessary skills to teach pronunciation properly to their students

It is common sense that everyone who has decided to learn a certain skill aims to master all components of it English learners are not different When they learn

Students learning English aspire to sound like native speakers or at least improve their intelligibility Dao's (2018) study demonstrates this aspiration and the common challenges they encounter Students desire proper pronunciation instruction to enhance their clarity and foster autonomous learning These findings align with previous research by Nguyen et al., highlighting the significance of effective pronunciation teaching in fostering language proficiency.

Both teachers and learners value proper pronunciation teaching and learning because only with intelligible pronunciation, students can be motivated to learn English and thus be able to produce what they have learned effectively This is not an easy task since other components have their place in a session as well If there is not much room left to squeeze in pronunciation instruction, then a different approach needs to be introduced to teach pronunciation effectively without adding more hours to the already stuffed schedules of the students

Giving feedback is crucial in helping learners improve their pronunciation However, as mentioned in the previous section, the lack of formal pronunciation instruction has led to reactive and intuitive ways of correcting leaners’ pronunciation mistakes Moreover, since teachers tend to correct learners’ pronunciation mistakes spontaneously, learners often feel that their mistakes are ignored by teachers (Plonsky & Mills, 2006) Hence, what would be an optimal way to correct pronunciation mistakes and how can all learners’ pronunciation mistakes be covered? This section is going to present possible answers for the preceding questions

Corrective feedback (CF) is an effective method for addressing pronunciation mistakes Studies by Ellis (2006), Lee et al (2015), and Kim & Han (2007) have demonstrated CF's effectiveness in improving learner pronunciation Specifically, CF has been shown to aid learners in correcting both lexical and phonological errors, enhancing their precision in pronunciation.

CF that pinpoint such errors Lyster et al., (2013) later explained the reason why

CF is helpful in addressing lexical and phonological errors is that these errors directly interfere with comprehension in the communication process in L2 Furthermore, CF is also considered as a type of explicit instruction which triggered the implicit learning process (Long, 2015) The importance of CF also lies in its ability to help learners deeply understand their mistakes and avoid making them later (Coupe, 2015) In general, CF leads to implicit knowledge, which then allows learners to improve their pronunciation outside the classroom (Ellis, 2009) Recast (i.e incorrect utterance is repaired/ restated), the most used CF technique, has been shown to be powerful in teaching pronunciation Saito & Lyster (2012) reported that recast helped adult Japanese English learners acquire the pronunciation of /ɹ/ Additionally, combining recast with explicit phonetic information would greatly enhance learners’ pronunciation comprehension

CF, or recast specifically, plays an important role in teaching pronunciation However, there exist obstacles in bringing it to pronunciation classrooms First, it is very difficult for teachers to practice recast with all students in a classroom Not only does it take time, but it also breaks the flow of a lesson when the teacher has to stop and do the correction whenever a student makes mistake This may affect students’ acquisition of other skills Second, teachers’ differences also pose another problem for students Since teachers have different accents, their recasts may confuse students Therefore, there calls for a solution of a continuous and standardized recast that learners can rely on This is when technology comes in, and further explanation will be presented in the coming sections.

A PPLYING TECHNOLOGY IN TEACHING PRONUNCIATION

The importance of pronunciation was stressed in the previous chapters However, teaching pronunciation is often neglected or done instinctively, based strongly on teachers’ intuition Moreover, the fact that teachers give one-on-one pronunciation feedback for learners is very time-consuming, not to mention that teachers’ accents may differ which may confuse learners For learning pronunciation to be effective, it needs to be done in “careful imitation rather than explicit instructions” (Newton,

2018) Furthermore, the types of feedback learners receive also affect their pronunciation acquisition Research has shown that learners greatly benefit from recast which is supported by an explicit phonetic explanation (Saito, 2013) Also, when learners could continuously repeat the corrected pronunciation, the improvement is even more noticeable A traditional pronunciation classroom may not guarantee all these ideal factors Fortunately, the advancement in the field of Computer-Assisted Language Learning (CALL) may meet all of the above requirements The emergence of Computer-Assisted Pronunciation Training or CAPT, A branch of CALL, has introduced new approaches to the field of teaching and learning pronunciation It has bridged the gaps that exist when pronunciation was instructed by teachers in traditional classrooms It offers learners with “endless opportunities for repetition and imitation, instantaneous response, and exposure to a wide range of target language speech” (Pennington & Rogerson-Revell, 2019) One key feature of CAPT is its on-going informative feedback, and to give such feedback CAPT programs base on automatic speech recognition (ASR) technology In the next sections, the researcher is going to discuss further on ASR and different types of feedback that CAPT programs offer and their advantages and disadvantages

2.3.2 Automatic Speech Recognition for Pronunciation teaching and learning

As the name suggests, with automatic speech recognition, or ASR, our speech is listened and analyzed by a machine for certain purposes For example, we can give commands to our phones by talking to it with Siri for iPhone, Google Assistant for Android phones, or Cortana for Windows laptops The machines listen to our speech, analyze, and execute the command accordingly ASR technology has also been applied in the field of teaching English pronunciation with a straightforward mechanism Learners record their utterances of words, phrases, or sentences to the CAPT program It then gives them feedback focusing on the phonological features of their utterance The types of feedback and functions may differ depending on the programs

One of the first and most notable program that used ASR in training pronunciation is Praat Developed by Paul Boersma and David Weenink of the University of Amsterdam, Praat analyzes a learner’s speech in terms of intensity, pitch height, duration, or formants, then compares the speech with a native speaker’s model through a display of speech waveforms Studies in applying Praat in teaching pronunciation have shown promising results Gorjian et al.’s study (2013) showed that Praat helped learners acquire prosodic features as well as stress and intonation in pronunciation skills More recently, in 2019, a study related to children’s speech showed that children’s literacy and pronunciation skills were improved after oral interactions with a pronunciation-based child robot (Spauling & Breazeal, 2019) Elsaspeak, an English-speaking practice app which uses speech recognition as its core function, also received positive feedback from learners, who reported that their pronunciation was significantly improved after using the app (Samad & Ismail, 2020) Although ASR-based CAPT programs have achieved remarkable results, they are not without limitations The selling-point in ASR-based CAPT programs, instantaneous feedback, also lie inherent drawbacks There are two main types of feedback in ASR pronunciation program, visual feedback, and ASR feedback With visual feedback, the learner’s utterance is displayed visually in a waveform graph (Fig 2.1) Programs such as Praat, Pronunciation Power offer this type of feedback However, it requires teachers’ specialized knowledge to interpret the results Moreover, it is doubted that learners can independently comprehend the feedback without teachers’ help In contrast to visual feedback, ASR feedback is straightforward and basic In programs such as Rosetta Stone, or Duolingo learner’s utterance is marked correct or incorrect, the extent of intelligibility is illustrated in a circle There is no specific analysis of phonological features in learners’ utterances

Despite the promising outcomes that CAPT programs have brought, there remain imminent demerits As Levis (2007) states “CAPT systems often suffer from difficulties in giving learners adequate, accurate feedback and an inability to provide an accurate and automatic diagnosis of pronunciation errors” (p.185) In the next section, the writer is presenting an ASR-based CAPT program that he used to teach pronunciation to his learners It is hoped to bridge the gaps that exist in previously mentioned programs.

S PEECH A CE

Created by Ian Cheung, Abhishek Gupta, and Ahmed El-Shimi, a group of ex- Microsoft veterans, SpeechAce is a Speech Recognition API (Application programming interface) for fluency and pronunciation assessment Its main feature is scoring a learner's speech and pinpoint individual syllable and phoneme level mistakes in a user's pronunciation in real-time SpeechAce API can be used to assess and score a variety of speaking activities:

1 Word and Sentence level pronunciation assessment

3 Estimating IELTS Speaking score and PTE Speaking score

4 Speaking activities (e.g Flashcards, fill in the blank, multiple choice)

5 Transcribing and evaluating spontaneous (free speaking) responses to a prompt

6 Evaluating the Grammar and Vocabulary level of a given text or response The SpeechAce API returns an overall score in each evaluated utterance and detailed sub-scores providing feedback on mistakes at sentence, word, syllable, and phoneme level

Figure 2 1 Praat waveform graph pronunciation feedback

SpeechAce’s function is straightforward and user-friendly for both the teachers who design the tasks, and the users doing the tasks First, teachers key in the prompt After that, users turn on the microphone on their electronic devices and record their answers SpeechAce then analyzes the speeches and gives feedback on users’ pronunciation SpeechAce can be integrated into various Learning Management Systems (LMS), such as Moodle (Fig 2.2), Blackboard, Canvas, thus teachers can customize the tasks depending on the speaking skills they want learners to practice Teachers can also listen to learners’ speech and add their own comments besides the comments from SpeechAce For users, they can practice their pronunciation endlessly and they can also have access to SpeechAce from different types of electronic devices

Users receive detailed feedback in two parts: a percentage score indicating utterance closeness to native speakers (0-100%) with corresponding comments ("you got it!") Additionally, they receive an analysis of individual syllables and phonemes, allowing users to compare their pronunciation with the model answer This comprehensive feedback helps users assess their pronunciation accuracy and identify areas for improvement.

Figure 2 2 SpeechAce task and feedback clicking on the text, a table drops down showing individual syllable and phoneme mistakes of users’ pronunciation Users can click on the specific word to listen and compare their speech with the model one (fig 2.3)

Besides giving feedback on users’ syllables and phonemes, SpeechAce also offers feedback in IELTS scores for speaking (ieltsace.com) With this feature, users are evaluated on their fluency and pronunciation Users record their answers in response to a question, and they receive their IETLS score accordingly They then can switch between the fluency and pronunciation tab to see their specific mistakes and what they are advised to do to improve their speaking performance (Fig 2.4)

Figure 2 3 Detailed analysis of pronunciation in SpeechAce

As can be seen from the two figures above, SpeechAce’s feedback is very familiar and easy to interpret even for beginner English learners Not only does it offer continuous practice, but it also notices learners with the present of phonemic features As a result, the learners will be aware of such existence and later they may teach themselves how to use this phonemic transcription to improve their pronunciation

SpeechAce is not the first program that uses ASR to teach pronunciation The pioneers in this area are Praat, pronunciation power, and later Rosetta Stone, Duolingo, and other similar programs These programs have brought positive outcomes in training pronunciation However, there exists drawbacks in their feedback demonstration For Praat and pronunciation power, the waveform graphs are difficult to interpret, for Rosetta Stone and Duolingo, the feedback is not specific enough to pinpoint learners’ mistakes Derwing et al (2015) criticized CAPT programs for their “one size fits all” approach He also pointed out other their limitation in offering diverse curriculum focus (phonemes), activity types (minimal pair distinguishing), feedback types (correct or incorrect) Consequently, according to Pennington (1999), for a CAPT program to meet the pedagogical requirements it should:

Figure 2 4 SpeechAce IELTS speaking feedback

1 Establish a baseline, reference accents for instruction

2 Set measurable goals and performance targets

3 Be designed to build skills from easier to more challenging exercises

4 Link pronunciation to other aspects of communication; and

5 Raise users’ awareness of how their L1 phonological systems differ from the system of the target language (p 434)

Besides the above requirements, Levis (2007) stressed that computer-based pronunciation feedback needs to:

1 be consistent with human feedback

4 be given in a form that students can make use of

5 include information about when goals have been reached; and possibly

6 suggest ways to address errors (pp 192–193)

With the previous description of SpeechAce’s functions and features, it seems that SpeechAce meets the criteria for an ideal CAPT program suggested by Pennington and Levis First, SpeechAce allows two versions of pronunciation models, the American and British accents Second, scores are shown in percentage to help learners set their goals more easily Next, SpeechAce also integrates with various LMS systems to help teachers customize the task freely Especially, SpeechAce’s pronunciation feedback is detail at the phonemes level which points out learners’ specific phonological mistakes Moreover, learners can use SpeechAce regardless of time and location with their smart devices, and they can also quickly get used to the user-friendly interface of the program Theoretically speaking, SpeechAce is an ideal CAPT program that teachers can rely on, as the developers claim “At SpeechAce we do one thing really well - We are the first and only speech recognition API designed for evaluating and giving feedback on pronunciation and fluency” (SpeechAce.com) Since its introduction, a number of studies have been done on SpeechAce’s effectiveness in improving learners’ pronunciation skills A study done in Italy in 2019 revealed that SpeechAce enhanced learners’ pronunciation skills as well as it boosted their self-confidence in speaking English

However, the authors of the study pointed out its drawbacks because of the lack of a control group (Aiello & Mongibello, 2019) Moxon (2021) reported that SpeechAce significantly improved the pronunciation accuracy of students Nonetheless, he admitted the limitations of the study due to the lack of male students and time constraints Lidia (2021) concluded in her study that SpeechAce has demonstrated its efficacy as a valuable resource for addressing English pronunciation challenges among secondary school students As we can see, the potential applications of speech recognition in pronunciation classrooms have called for further research on programs such as SpeechAce As a result, the author has decided to carry out a similar study to those of Aiello, Mongibello, and Moxon This study should bridge the gap of the previous studies due to its inclusion of a control group and an experimental group, and an equivalent number of male and female participants The study hopes to reveal positive results on using SpeechAce to improve learners’ pronunciation and from that, the author calls for further research to be done on the topic of application ASR in pronunciation classrooms to make learning and teaching pronunciation more feasible for both teachers and learners

Although limited in empirical studies on its effectiveness, what SpeechAce offers have great potential in the field of training pronunciation As mentioned in previous parts, for pronunciation teaching and learning to be effective, learners need to be able to practice as much as possible and be exposed to a standardized pronunciation model Moreover, learners’ knowledge of phonological features such as the ability to interpret phonemic transcription is also crucial in developing pronunciation competence SpeechAce guarantees those requirements The rest depends on how teachers apply SpeechAce in their pronunciation teaching.

RESEARCH METHODOLOGY

R ESEARCH D ESIGN

To answer the two research questions, the study applied a mixed-method experimental design Firstly, a mixed-method approach is particularly well-suited for educational research, as it allows for a comprehensive understanding of both the quantitative outcomes and qualitative experiences of participants (Creswell & Plano Clark, 2017) This approach can yield rich, detailed data that provides a more complete picture of the phenomena under study By combining quantitative methods (like pretests and posttests to measure changes in pronunciation competency) with qualitative methods (such as interviews or focus groups to explore student perceptions and experiences), the researcher can examine not only the efficacy of SpeechAce but also the contextual factors influencing its use and effectiveness The experimental design element of the research, involving control and experimental groups, is crucial for establishing a causal relationship between the use of SpeechAce and improvements in pronunciation competency Experimental designs are considered the gold standard in educational research when seeking to determine cause-and-effect relationships (Shadish, Cook, & Campbell, 2002) By comparing outcomes between groups that use SpeechAce and those that do not, the researcher can more confidently attribute any observed differences in pronunciation competency to the intervention itself Furthermore, the choice of a mixed-method approach aligns well with contemporary educational research, which increasingly recognizes the value of integrating both quantitative and qualitative data to address complex research questions (Greene, 2007) This integration allows for a more nuanced understanding of how and why SpeechAce may be effective, or conversely, where it may fall short

The mixed-method experimental design was chosen for this study due to its ability to combine the rigor of experimental methods with the depth of qualitative insights This combination allows for a holistic evaluation of SpeechAce's impact on pronunciation competency among non-English majors at Hoa Sen University.

Now that the designed have been determined, the researcher needed to decide on the two groups for the study To randomly assign two groups for the study would disrupt the university’s training schedule As a result, convenience sampling with two intact groups would be chosen for the study While convenience sampling may not provide a representative sample of the broader population, which can limit the generalizability of results, its practicality and cost-effectiveness make it an appealing option for many studies, particularly in the fields of social sciences and education (Creswell, 2014).The researcher chose two of the three classes he was in charge at the time as control group and experimental group Procedurally in experimental design, pretest and posttest will be applied to measure the difference in performance of the targeted skill, in this case, pronunciation competency between the two groups Drawing from the previous description, research question

1 will be answered with an experimental design

The second research question aims to gauge students' perceptions of pronunciation learning using SpeechAce through both a questionnaire and interviews The questionnaire provides a broad understanding of these perceptions, while interviews delve deeper into students' experiences with SpeechAce This research question necessitates a mixed-method approach that combines quantitative (questionnaire) and qualitative (interviews) data, making a mixed-method design a suitable option.

In conclusion, to answer both research questions, a mixed method quasi- experimental design will be viable The design is illustrated in table 3.1 below

Posttest Select Experimental group Treatment Questionnaire

Table 3 1 Mixed method quasi-experimental designs

R ESEARCH SITE

Hoa Sen University, established in 1991, hosts the Department of General English, part of its esteemed Faculty of Foreign Languages The English for International Communication (EIC) program, offered by this department, aims to equip students with English proficiency across various contexts, including multicultural workplaces and academic settings Progressing through six levels (EIC 1 to EIC 6), students demonstrate their proficiency by scoring 5 or above in each level The program culminates with the Aptis test developed by British Council, requiring a score of 120 for successful completion English File 3rd edition (Oxford University Press) serves as the primary course book, supplemented by additional materials tailored to specific levels.

Course books Level – CEFR Duration EIC

Intermediate plus Unit 6 – Unit 10) Inter plus (2) - Mid

Intermediate plus (Unit 1 – Unit 5) Inter plus (1) - Mid

Intermediate (Unit 1 – Unit 5) Inter (1) - B1 105 periods

English File (3 rd edition) - Pre-

Table 3 2 English for International Communication (EIC) study path

P ARTICIPANTS

The participants of the study are from two EIC 4 classes coded EIC4 – 0500 and EIC4 – 0200 As mentioned previously in the research design section, the two classes are chosen as intact groups for the study, in which, class EIC4 – 0500 (n7) is the experimental group and class EIC4 – 0200 (n7) is the control group Table 3.3 provides further information on the participants

Group CG (n7) EG (n7) Total (N) Percentage (%)

Both groups are at level 4 of the program, in which they use the book English file – Intermediate (unit 6 to unit 10) All participants study three 2.5-hour sessions a week The researcher was in charge of the Thursday and Saturday session The foreigner co-teacher taught the Tuesday session The participants’ assessment components are homework score, attendance record, unit tests, project, midterm test, and final test Table 3.4 displays detailed assessment scheme for the level

Level Progress Assessment Midterm Final

(Speaking: 20; Writing: 20; Reading, Listening, Vocabulary & Grammar: 60 Total score (100) will be converted to a 10-grading scale.)

The participants in both groups will receive the same treatment in all lectures except for the pronunciation lessons In this aspect, the participants in the control group will learn pronunciation from the book English file – Intermediate (3 rd edition, by OUP), from unit 6 to unit 10 Meanwhile, the participants in the experimental group will learn the same pronunciation content from the same book, but on SpeechAce platform In other words, the pronunciation tasks on the book is converted into SpeechAce form for the participants in the experimental group to practice The description on how the paper-based pronunciation is converted into SpeechAce form will be discussed in the coming section.

R ESEARCH INSTRUMENTS

The instrument section of this thesis outlines the tools used to collect data and measure the effectiveness of SpeechAce software in developing pronunciation competency for non-English majored students The following subsections describe the instruments used in this study

As mentioned earlier, participants in the experimental group will undergo pronunciation training with SpeechAce This section outlines the conversion process of pronunciation content from the "English File - Intermediate" textbook into the SpeechAce app.

To replicate the exact paper-based pronunciation task into SpeechAce is impossible As a result, the researcher made use of all available features in SpeechAce to maintain the same characterisctic between the SpeechAce-based pronunciation task and paper-based pronunciation task Fortunately, SpeechAce is designed to integrate with the LMS system Moodle, hence there are more available features that can be adapted into SpeechAce to maintain its similarity with the paper-based pronunciation task A comparison between the paper-based and SpeechAce-based pronunciation tasks for control group and experimental group is presented as below table

Task Listen and complete the sentences Play the audio file - fill in the gap play model utterance - record one's own utterance - review SpeechAce's score & analysis

If I _ more , I'd be a lot healthier

Task Circle the word with different sounds read each word aloud - circle record one's own utterance of all words

Question bite - smile - height - weight

Task Listen and write sentences play the audio file - write down the sentence play model utterance - record one's own utterance (sentence appears after students have recorded their utterance) - review SpeechAce's score & analysis

"I'm going to write an email to complain"

Table 3 5 Pronunciation teaching procedure for control group and experimental group

SpeechAce's model utterances are all in a British accent, aligning with the textbook's pronunciation model Students can attempt pronunciation exercises on SpeechAce indefinitely Upon receiving a student's utterance, SpeechAce provides a detailed analysis of their pronunciation (as depicted in Figure 3.1).

3.4.2 The pre-test and posttest

Prior to the treatment phase, both groups take the pronunciation pretests (Appendix A) The questions of the pretest are taken from the book test your pronunciation by Michael Vaughar (Vaughar, 2002) The test consists of seven tasks with a total of 37 questions which test students’ competence on key areas of English pronunciation such as identifying vowels and consonants, the relationship between sounds and spelling, using, and understanding phonemics transcript, sentence

Figure 3 1 Interface of SpeechAce’s analysis of learners’ pronunciation stress and intonation Different types of tasks are integrated most of which are multiple choice questions Students are also required to listen and fill in the gaps for weak/ strong form exercise and write words from phonemic transcription

The same procedure is repeated for the posttests (Appendix B) The questions, however, are taken from another textbook, the English pronunciation in use by

Mark Hancock (Hancock, 2003) The test also covers the key areas in English pronunciation which were tested in the pretest Students are required to identify sounds, stress pattern, intonation, syllable, and phonemic transcript They are also required to do some listening and writing which are similar to the pretest The number of questions, however, doubles that of the pretest with a total of 74 questions The purpose of this significant increase in number of questions is to gauge students’ pronunciation competency in an in-depth manner

The rationale behind the inclusion of sounds, stress patterns, intonation, syllables, and phonemic transcription in the pretest and posttest is that they are essential for assessing English pronunciation competency, as these are key aspects of English phonology The accurate production of phonemes (sounds) is foundational, and mispronunciations can lead to significant communication issues (Celce-Murcia, Brinton, & Goodwin, 2010) Stress patterns are critical in English, a stress-timed language, where incorrect stress can change meanings (Underhill, 2005) Intonation, or the rise and fall of voice pitch, conveys meaning and emotion, essential for effective communication (Brazil, 1997) Syllable mastery is also vital due to English's complex syllable structure, impacting spoken communication clarity (Roach, 2009) Additionally, phonemic transcription, representing sounds

Incorporating visual aids into pronunciation tasks aids ESL learners in discerning and practicing sounds, especially those not present in their native language By doing so, researchers not only assess pronunciation but also equip learners with tools and knowledge essential for improving their English pronunciation skills effectively.

Both pretest and posttest are adopted from English pronunciation testing and training textbooks with distinct answer keys, as a result, inter rater would not be necessary The researcher, therefore, act as a sole rater who scores the tests by using answer keys taken from the textbooks

At the end of the study, participants from the experimental group complete a questionnaire which is designed to measure their perceptions toward learning pronunciation as well as using SpeechAce to improve their pronunciation competency Ten volunteer students also take part in interviews to further elaborate their perception in the same issues in the questionnaire

The questionnaire, formulated on a five-point Likert scale ranging from "strongly disagree" to "strongly agree," encompasses three sections with a total of 23 items

It aims to assess students' perceptions and challenges in learning pronunciation,

Figure 3 3 A task in the posttest both prior to and subsequent to their experience with SpeechAce The first section, focusing on perceptions of pronunciation, derives its items from Nowacka's (2012) study, which similarly explored learners' views on English pronunciation learning This adaptation was chosen due to the alignment of Nowacka's research objectives with the current study's aim to gauge learner perceptions The remaining items in the questionnaire were specifically developed by the researcher, crafted to anticipate and evaluate the potential outcomes of incorporating SpeechAce into pronunciation training These items were formulated based on the researcher's expectations and understanding of the tool's impact The questionnaire, presented in English, was administered to students immediately following the posttest to capture their immediate responses and reflections

Questionnaire sections Number of items

According to Tuckman (1972) an interview could be employed as the primary method for collecting data that fulfills the research goals, capturing insights into an individual's thoughts, knowledge, preferences, values, and beliefs Therefore, to further understand the learners’ perspective towards using SpeechAce in pronunciation learning, an interview is initiated after the participants of the control group have finished the questionnaire Ten volunteer students are invited to do the one-on-one interview in a separate room The other students wait outside for their turn This is to maintain the objectivity of the interview and the next interviewee’s answers will not be influenced by the preceding one

The interview consists of 9 questions (Appendix D) which are adapted from the questionnaire to give participants chance to further elaborate their answers in the questionnaire as well as to give the author a broader view on the effectiveness of

D ATA COLLECTION PROCEDURE

The course lasts 13 weeks The first week is for the author to gain access and carry out the pronunciation pre-test for the two groups The next 11 weeks, from week

2 to week 12, are dedicated for the treatment process During this stage, the control group is taught pronunciation solely from the course book, while the experimental group learn pronunciation with SpeechAce Week 12 is for the posttest, questionnaire, and interview The participants in the two groups take the posttests The participants in the experimental group do questionnaire and interview Week

13 is for results phase, during which the data collected from the research instruments is analyzed and interpreted The schedule is planned specifically as the table below:

Gain access Email the head of the department to obtain permission to do research

Pre-test for experimental group and control group

Run independent t-test to compare pronunciation score

Pronunciation lesson from course book for control group

3 Result discussion Posttest for control group and experimental group 13

R ELIABILITY AND VALIDITY

Creswell (2012) defines reliability and validity in research as follows: reliability refers to the consistency of results across time, instruments, and participants, while validity means that the findings accurately represent the population being studied

To guarantee such reliability and validity of the study the notion of triangulation is applied The social sciences use triangulation methods to gain a more comprehensive understanding of complex human behavior by studying it from multiple perspectives and using both qualitative and quantitative data This approach helps to better explain the richness and complexity of human behavior Triangulation is a valuable method for demonstrating concurrent validity (Cohen, Manion, & Morrison, 2018, p 330)

The study is triangulated as the table below:

Pronunciation pretest and posttest for students of both groups

Questionnaire for participants of experimental group

SpeechAce improve students’ pronunciation competency?

What are the students’ perceptions toward

Questionnaire and interview for experimental group

Table 3 7 Outline of data collection procedure learning pronunciation with

Table 3 8 Outline of triangulation method

D ATA ANALYSIS SCHEME

This section describes how the data collected from the pretest, posttest, and interviews from the experimental group and control group is analyzed

3.7.1 Analysis scheme for the pretest and posttest

To analyze the result of the pretest and posttest, the author used the Statistic Package for the Social Sciences (SPSS) SPSS provides a user-friendly interface for data analysis, with a wide range of statistical procedures and tests available to researchers It allows users to input and manipulate data, create custom variables, generate descriptive statistics, and conduct various tests such as t-tests, ANOVA, regression analysis, and factor analysis SPSS also provides the ability to create graphs and charts to visually display data It is a widely used tool in social science research due to its flexibility, ease of use, and powerful analytical capabilities Therefore, SPSS would be an optimal tool to analyze the data of the test

First, mean scores of pretests and posttests were calculated to see whether the average scores of both groups are similar However, mean alone does not provide information on the variability or spread of the data As a result, standard deviation values were also calculated to see how spread out the data is from the mean A small standard deviation indicates that the data points are clustered around the mean, while a large standard deviation indicates that the data points are more spread out Next, an independent sample t-test was used to determine whether there is a significant difference between the means of two groups and conclude the effectiveness of the interventions

3.7.1 Analysis scheme for questionnaire and interview

A questionnaire was utilized to ascertain the participants' perceptions of practicing pronunciation with SpeechAce The questionnaire comprised questions that adhered to a uniform format, featuring a five-point scale from "strongly disagree" to "strongly agree." The data collected from the questionnaire was processed and analyzed utilizing the SPSS The Cronbach Alpha of the questionnaire, in addition to each dimension of the issue mentioned in the questionnaire, was computed Additionally, a descriptive analysis was conducted to determine the mean, frequency, and percentage of each item and dimension (a group of items) Based on the results of the descriptive analysis, the researcher could arrive at a conclusion regarding the issue under investigation

Thematic analysis, a qualitative research method, is widely employed to identify and analyze patterns (themes) within a dataset By uncovering underlying meanings and concepts, thematic analysis offers valuable insights into experiences, perspectives, and beliefs This technique empowers researchers to delve into complex issues and capture diverse viewpoints, contributing to theory development, policy-making, and future research in various fields, including social sciences and psychology.

The process of thematic analysis involves several key steps First, researcher revise the interview questions carefully to ensure the questions meet the aim of the research Next, researchers start identifying meaningful units within the data and label them with codes After coding, researchers group related codes together to form themes, patterns that represent significant concepts or ideas present in the data Researchers then explore how themes are interconnected and examine the relationships between different themes In the final step, researchers report the findings through a written narrative that describes the themes, supported by relevant quotes from the data.

FINDINGS AND DISCUSSION

T O WHAT EXTENT DOES S PEECH A CE IMPROVE STUDENTS ’ PRONUNCIATION COMPETENCY ?

The data collected to answer this research question includes the pronunciation scores of the pretest and posttest of the experimental group and control group

4.1.1 Independent Sample T-Test of pretests results

First, to ensure both groups has the same starting point in terms of their pronunciation competency, the two groups take the pronunciation pretest The mean scores of the pretest from the groups are then analyzed using independent sample t-test to determine if there is a significant difference between them

From the table, the mean scores of the pretests for CG (Control Group) and EG

(Experimental Group) were computed Table 4.1 indicates that the CG pretest mean score is 30.93, while the EG pretest mean score is 31.47 There is a slight difference between these two values, with the mean score of CG (M0.93,

SD=5.948) being marginally lower than the mean score of EG (M1.47,

SD=4.299) To determine whether this difference is statistically significant, a test was conducted for further examination

Levene's Test for Equality of Variance s t-test for Equality of Means

95% Confidence Interval of the Difference

Std Error Mean Pretest pronunciation score

Table 4 1 Descriptive statistics of pretest scores

Upp er Prete st score

Table 4 2 Independent Sample T-Test of pretests results

As observed in table 4.2, the outcomes of the Independent T-test were presented

Initially, the Levene's test was conducted to assess the equality of variances, and the Sig value obtained was 167 (Sig > 05) This indicates that the conditions for equal variance were met Subsequently, the T-test for equality of means was performed, and the Sig (2-tailed) value obtained was 671 (> 05) Consequently, the differences between the means of CG and EG were found to be statistically insignificant, implying that they are considered equivalent In conclusion, before the treatment, the pronunciation competency of both the CG and EG was comparable to each other

4.1.2 Independent Sample T-Test of posttests results

The procedure to analyze pretest pronunciation results was repeated for the results of the posttest First, mean scores of the two groups are calculated in table 4.3

As reported from the table 4.3, the control group's mean score is 51.80 with a standard deviation of 5.839, while the experimental group's mean score is 58.03 with a standard deviation of 4.687 Thus, EG’s mean score is considerably higher than CG’s mean score Afterward, a test will be performed to determine if the disparity between the mean scores of the control group (CG) and the experimental group (EG) holds statistical significance

Independent Samples Test Table 4 3 Descriptive statistics of posttest scores

Levene's Test for Equality of Variances t-test for Equality of Means

95% Confidence Interval of the Difference Lower Upper Posttes t Score

Table 4 4 Independent Sample T-Test of posttests results

Table 4.4 reveals equal variances between groups, confirmed by a non-significant Levene's test (Sig = 257, > 05) This permits the assumption of equal variances, allowing for further statistical analyses.

.00, which is lower than 05 This finding confirms the existence of a statistically significant difference between the posttest means of the control group (CG) and the experimental group (EG) Consequently, the pronunciation competency of both groups changed after the treatment, with the performance of the experimental group being notably better than that of the control group.

W HAT ARE THE STUDENTS ’ PERCEPTIONS TOWARD LEARNING PRONUNCIATION WITH

The questionnaire is delivered to the students of the experimental group at the final session

It is used to gauge students’ attitude and difficulties in pronunciation as well as their views towards adapting SpeechAce into their pronunciation learning

Initially, there were 24 items in the questionnaire, however, after the pilot test, one item was removed to satisfy the reliability and validity of the questionnaire The questionnaire now consists of 23 items, divided into three sections

The first section consists of 6 items to measure student’s attitudes toward pronunciation and 1 item to survey students’ preference in learning pronunciation

In the second section, students express their difficulties in learning pronunciation through a set of 7 items The final section includes 8 items in multiple choice Likert scale form and one open-ended question It surveys students’ attitude towards using SpeechAce to learn English pronunciation The development of the questionnaire was based on a five-point Likert scale, spanning from (1) strongly disagree to (5) strongly agree

To ensure questionnaire reliability and internal consistency, Cronbach's Alpha analysis was conducted for its three sections using SPSS version 28 The results are presented in three tables.

Cronbach's Alpha Based on Standardized Items N of Items

Table 4 5 Reliability statistics of questionnaire

According to Bryman and Cramer (1990, p.71), the level of reliability is deemed satisfactory at Cronbach’s Alpha of 0.8, while alternate viewpoints propose acceptability if it reaches 0.67 or higher From the three tables presented earlier, it is evident that all the results from the analysis of data collected yielded Cronbach's alpha values of 0.708, 0.723, and 0.755, all of which are slightly above 0.67 This indicates that the questionnaire exhibited an acceptable level of reliability Consequently, the obtained results provide evidence that the questionnaire instrument is both valid and dependable

Now that the questionnaire meets the requirements of reliability and validity, each section is further interpreted to find out the patterns in the students’ response

4.2.1.1 Students’ perceptions and difficulties in learning pronunciation

Figure 4 1 Students' perceptions towards pronunciation

As can be seen from the figure 4.1, all students agree that possessing good English pronunciation is important for them, with 19 students (59.38%) strongly agree with this statement and 13 students (40.63%) agree with this Most students express their enthusiasm in learning pronunciation with half of the students surveyed found it interesting to learn pronunciation, while only 2 students chose neutral From the first two items’ results, students have very serious attitude towards learning English pronunciation, not only did they find it is an important component to acquire, but they also found interest in learning it Most students target for native- like pronunciation with more than half of the participants agree with this item and nearly half of them strongly agree with this, while only 2 students express a neutral stance and none of the surveyed participants strongly disagree or disagree with this item The result in item 3 seems to synthesize with that of item 4, which ask participants to confirm whether they think the optimal way to perfect English pronunciation is by speaking to native speakers The result shares similar pattern with item 3’s results, with nearly 90% of the participants agreed or strongly disagreed, whereas no participants disagreed or strongly disagreed with this assertion In item 6, a considerable number of participants viewed it possible to achieve native-like pronunciation Only 1 student disagreed with this statement, meanwhile 8 of them took a neutral stance and the rest of the group either agreed or strongly disagreed It seems that students believe it is possible to speak English like a native speaker with proper practice and learning technique However, item 5’s result has shown that students do not receive on-going pronunciation learning, with 24 participants chose neutral when being asked to confirm if textbooks are their main sources of learning pronunciation

Listening to authentic English is useful for learning pronunciation 18.8% 81.3%

Attending Phonetic classes is useful for learning pronunciation 53.1% 46.9%

Speaking with native speakers is useful for learning pronunciation 21.9% 78.1%

Imitating authentic speech is useful for learning pronunciation 43.8% 56.3%

Study by yourself is useful for learning pronunciation 25.0% 75.0% Learn pronunciationrning phonemic transcription is useful for learning 18.8% 81.3%

Table 4 6 Students' perceptions in learning pronunciation

The first section of the questionnaire also surveys students’ perceptions in learning pronunciation As can be seen from table 4.8, most students viewed that it is useful and effective to be exposed to authentic English speech In addition, 81.3% of the participants also agreed that learning the phonemic transcription is useful for their pronunciation They also expressed a preference for doing self-study when they learn pronunciation

Students’ difficulties in learning pronunciation

Figure 4 2 Students’ difficulties in learning pronunciation

The next section further explored students’ difficulties in learning pronunciation

More than half of the participants (17 students) agreed that they did not know how to learn pronunciation by themselves, 3 students strongly agreed with this On the other end of the spectrum, 6 students disagreed with this statement, and the other

6 expressed a neutral view Being able to comprehend phonemic transcription is the key point in self-learning pronunciation, this view is surveyed in item 8, “I cannot read phonemic transcriptions” Nearly half of the group confirmed this view

(12 students agreed and 3 disagreed) Also, students also found it difficult to find native speakers to talk to, with 13 students (40,6%) agreed and 7 students (21.9%) strongly agreed with the statement “I cannot find native speakers to practice my pronunciation” Ending and linking sounds are two prominent features in English pronunciation which are troublesome to most Vietnamese English learners This assumption was proven in the participants’ responses to item 10 and 11 The same number of participants, 10 students, agreed with these two items Moreover, 6 students chose strongly agree for difficulty in linking the sounds together and 5 students picked strongly agree for difficulty in producing the ending sounds

Difficulties in learning pronunciation exist, however, it is not because teachers spend little time teaching students pronunciation More than two third of the students were on the strongly disagree – disagree spectrum when being asked to confirm the statement “teachers spend little time in teaching pronunciation”

“Learning pronunciation from different teachers’ accents confuses me” shares a mixed views from students, with the same number of students, eight of them, agreed and disagreed with this item, while “strongly agree” and “neutral” were chosen by 6 students for each

Figure 4 3 Students’ perceptions towards SpeechAce

Now that students’ attitude and difficulties in pronunciation are gauged, SpeechAce comes into play hoping to bridge the existing gaps in achieving good pronunciation In the first two items, no students disagreed or strongly disagreed with the fact that SpeechAce is easy to use, and it has helped them improve their pronunciation Specifically, 31 over 32 students (96.88%) surveyed deemed SpeechAce easy to use, only one ticked neutral In addition, 17 students agreed (53.13%), and 11 of them (34.38%) strongly agreed that SpeechAce did help them improve their pronunciation, only 4 students chose neutral SpeechAce has also somewhat increased students’ ability in reading phonemic transcription, with 14 students agreed and 6 students strongly agree with item 16 Meanwhile, there were also 11 students showed no position in this item, and only one of they disagreed that SpeechAce can help them read phonemic transcription However, SpeechAce has considerably helped students improve their ending sound missing error More than half of the group either agreed or strongly agreed with this, higher than the number of students who chose neutral and disagree Students have also benefited from detailed analysis of their pronunciation of SpeechAce, with all of the students, except one, either agreed or strongly agreed with item 20 Moreover, a relatively high number of students reported that they can imitate the model pronunciation from SpeechAce, with two third of them fell on the agree-strongly disagree spectrum Additionally, students can also learn SpeechAce effectively on their own, with 18 of them agreed and 7 strongly agreed with item 19, the other 7 students showed no position with this statement

Speechace has proven effective in enhancing pronunciation skills among learners Participants expressed a preference for using Speechace in pronunciation classes due to its engaging, practical, and user-friendly interface The detailed analysis of pronunciation mistakes enabled learners to identify improvement areas and ultimately enhance their performance Overall, participants' positive feedback highlights the significant impact of Speechace in fostering pronunciation proficiency.

Individual semi-structured interviews were conducted with ten volunteer students (eight females, two males) in a classroom setting The interviews lasted four to five minutes each and were audio-recorded with participant consent Each participant received the same nine questions following a structured order To protect anonymity, participants were assigned codes from S1 to S10 The interview questions were grouped into three sections: students' background and pronunciation learning challenges, their perspectives on using AI for pronunciation, and their perceptions of using SpeechAce for pronunciation learning.

4.2.2.1 Students’ background and their difficulties in learning pronunciation

When being asked about how they have been learning pronunciation previously, and whether they find them effective, the students shared similar answers Most of them said that they learn pronunciation mainly at schools or English language centers One student had chance to learn pronunciation with native speakers (S5) Others learned from the internet, such as watching short clips on YouTube or Facebook, watching movies, listening to music (S7, S8, S9) Four students also mentioned the apps Cake and TFLAT (S4, S8, S9, S10) as their pronunciation learning tools However, when it comes to the effectiveness of their pronunciation learning process, most students expressed uncertainty about the improvement of their pronunciation S9 said that “it only has short-term effectiveness However, I can’t recall what I learned later on” Similarly, S5 deemed her pronunciation learning effective, but she does not think it has been improved S3 stated that “the pronunciation teaching methods at school may work for students with aptitude, but not for me” In terms of the difficulties they have in learning pronunciation, there were four main ones Four students reported that different methods of teaching pronunciation from different teachers confuse them

D ISCUSSION

In this chapter, the data presented previously is further discussed and synthesized with previous studies mentioned in literature review section It aims to see how thoroughly the collected data answers the two research questions Thus, it is divided into two parts The first part bases on the collected data from pretest and posttest to discuss students’ improvement on pronunciation competency after using SpeechAce The second part combines data collected from questionnaire and interviews with experimental group to conclude on students’ perceptions towards using SpeechAce to practice their pronunciation

4.3.1 To what extent does SpeechAce improve students’ pronunciation competency?

The pretest results show that EG's mean score was slighly higher than that of CG’s mean score with 31.47 and 30.93 respectively (table 4.1) However, the Sig (2- tailed) value obtained from the independent samples test was 671 (> 05) (table 4.2) Thus, it can be concluded that, the differences between the means of CG and

EG were relatively insignificant After eleven weeks of applying SpeechAce to the

EG, the posttest results show that EG’s mean score is now relatively higher than CG’s mean score, at 58.03 compared to 51.80 (table 4.3) Moreover, the The 2- tailed Sig value was found to be 00 (table 4.4), which is below the significance level of 05 This result indicates a statistically significant difference between the posttest means of the control group (CG) and the experimental group (EG) Therefore, it can be confirmed that the pronunciation competency of both groups improved after the treatment, with the performance of the experimental group notably surpassing that of the control group This finding parallels with the results from the three studies previously mentioned in the literature review section by Aiello and Mongibello (2019), Moxon (2021), and Lidia (2021) However, Aiello and Mongibello acknowledged the lack of control group in their study, and Moxon

He admitted the study's restrictions stemming from the absence of male students and limitations in terms of time This study, which was done with both control group and experimental group during a process of 10 weeks, has bridged the gaps mentioned by previous authors All of these factors validated the impact of improving students’ pronunciation competency through the utilizing of SpeechAce

4.3.2 What are the students’ perceptions toward learning pronunciation with SpeechAce?

Before the study, the researcher thought that students would take learning pronunciation for granted as a few of his students have fairly unintelligible pronunciation, and those students do not bother practicing sharpening their pronunciation Moreover, from all his years of teaching, he has only taught a few pronunciation sessions separately Instead, most of his pronunciation teaching occurs spontaneously in an uninformed manner However, the data obtained from the questionnaire has surprised the researchers All students have shown that they have very serious attitude towards learning pronunciation, specifically, all students either agreed or strongly agreed that “it is very important to possess good pronunciation”, and most of them also found it interesting whenever they got to learn pronunciation, such positive attitude was unexpected from the researcher’s point of view Not only that, but most students also held a very optimistic view towards their ability to learn a native-like accent in pronunciation, this attitude from the students matches with the study done by Dao (2018)

Unfortunately, such enthusiasm and positivity in learning pronunciation of the students were hindered by the difficulties they had to face In fact, many students’ difficulties matched with previous points mentioned in the literature review chapter Specifically, nearly two thirds of the surveyed participants admitted that they did not know a proper method to learn pronunciation by themselves even though many of them disagreed or strongly disagreed that “teachers spend little time in teaching pronunciation This is likely because the majority of instructors taught pronunciation in methods that were not premeditated and were characterized by reacting to immediate needs (Nguyen & Newton, 2020) This difficulty was also mentioned by four participants during the interview (S1, S2, S3, S4) In previous chapter, the researcher assumed that most Vietnamese English learners had troubles in recognizing and pronouncing ending and linking sounds Certain English phonetic sounds, including t, ʧ, ʤ, ʒ, s, θ, and ʃ, pose challenges for learners (Tam, 2005) These findings are supported by the results from the questionnaire, where half of the students agree or strongly agree with these assumptions, and by three interviewees (S1, S7, S8) Moreover, many students were not not particularly familiar or confident with using phonemic transcriptions in learning pronunciation

The students highly valued SpeecAce's ease of use and its potential for effective self-directed pronunciation learning Almost all participants believed they could accurately identify their pronunciation errors and imitate the correct sounds, suggesting that SpeecAce effectively addresses the challenges of phonemic transcription interpretation and ending-sound omission Additionally, SpeecAce provides detailed feedback, allowing students to pinpoint errors and make targeted corrections, a key aspect of effective pronunciation instruction.

& Han, 2007; Lee et al., 2015; Long, 2015; Lyster et al., 2013; Saito & Lyster, 2012) Undoubtedly, SpeechAce has offered the appropriate form of corrective feedback that learners require to enhance their pronunciation learning with greater effectiveness Finally, most students admitted that their pronunciation had improved after using SpeechAce, all participants in the interview also supported this view and half of the surveyed participants expressed that SpeechAce could help them pronounce like native speakers These perceptions agree with what previously presented in the literature review

It seems that SpeechAce had broken all the barriers that stop students from learning pronunciation effectively Also, it has helped them improve their pronunciation in certain areas However, it should be noted that there exist some uncertainty amongs students’ answers Thus, it would never be agood idea to rely solely on SpeechAce to teach pronunciation, as many of the interviewed students reported that they still prefered real teachers to teach them pronunciation because it is not only the proper pronunciation they are looking for, they are also wish to talk with real people to boost their confidence in communication All in all, it would be wise to use SpeechAce only as an effective tool to help teachers bridge the gaps in teaching pronunciation to students, and the idea of A.I that can completely replace a human teacher has yet to come.

CONCLUSION

S UMMARY OF FINDINGS

The study aims to investigate the impact of SpeechAce in developing pronunciation competency for non-English-majored students at Hoa Sen University It seeks to answer the two questions: “to what extent does SpeechAce improve students’ pronunciation competency?” and “what are the students’ perceptions toward learning pronunciation with SpeechAce?” A mixed method quasi-experimental design was applied with one control group (CG) and one experimental group (EG) The research tools were one pretest for both groups, one questionnaire and interview sessions for EG These tools serve the purpose of triangulation and increase the reliability and validity of the study

The results from the pretest and posttest have shown that students pronunciation competency of the EG was better than that of CG As a result, the effectiveness of using SpeechAce to improve pronunciation was confirmed The data obtained from the questionnaire and interviews have revealed that the majority of students received SpeechAce in very a positive manner although there lie some uncertainties which calls for teachers’ careful consideration and preparation when using SpeechAce to teach pronunciation.

I MPLICATIONS

As teachers, it is noteworthy that students are generally enthusiastic about learning pronunciation, exhibiting a positive attitude towards this aspect of language acquisition This finding aligns with Gilbert (2008), who emphasizes the importance of capitalizing on students' willingness to learn pronunciation Consequently, teachers need to adjust their approach to teaching pronunciation Despite the challenges, such as the time-consuming nature of pronunciation instruction, teachers have a critical role in making pronunciation learning meaningful and engaging for students Derwing and Munro (2005) highlight the necessity for teachers to not only instruct but also model accurate pronunciation to facilitate students' acquisition Additionally, teachers must focus on correcting common pronunciation errors specific to Vietnamese English learners, as indicated by Jenkins (2000) Incorporating phonemic transcription into teaching, as suggested by Celce-Murcia et al (2010), is crucial, as it has been proven in this study to significantly benefit students’ learning process Above all, embracing AI technology, such as SpeechAce, and integrating it into teaching practices can enhance the effectiveness of both learning and teaching, as recommended by Chapelle (2001)

For students, the provision of SpeechAce offers a valuable tool for independent pronunciation practice However, to fully utilize this tool, learners must become proficient in interpreting phonemic transcriptions, a skill that Levis (2005) notes is essential for effective pronunciation learning This also requires a level of self- discipline for consistent self-study

Integrating AI technology in English curriculum, as evidenced by SpeechAce's positive impact, supports Pennington's (1996) advocacy for technology integration Moreover, formal pronunciation training for educators enhances their confidence and competence, benefiting students As Derwing and Munro (2015) emphasize, teacher confidence in pronunciation instruction directly impacts student outcomes.

L IMITATIONS

Although the study has proven what aims to find, it is essential to acknowledge certain limitations that may influence the interpretation of the findings The first limitation is in the sample size The study was done with 74 participants in total The study's sample size was constrained by practical considerations, potentially limiting the generalizability of the findings to broader populations Additionally, the study focused on a specific demographic, and the impact of SpeechAce on learners from diverse linguistic backgrounds might differ The researcher believes that a larger sample size would increase the reliability and validity of the result The participants' familiarity with technology, particularly ASR-based tools like SpeechAce, could have influenced their engagement and outcomes Learners with varying levels of technological proficiency might have experienced different levels of comfort Pronunciation development is influenced by multiple factors, including individual learner characteristics, learning styles, and prior language exposure The study did not extensively explore these interactions Another limitation lies within the transformation of pronunciation exercises from textbook into SpeechAce form There were some features which have yet to exist in SpeechAce which makes the transformation from book to SpeechAce inaccurate, hence it may affect the validity of the activity The design of pretest and posttest also raise some concerns None of the tasks in the pronunciation pretest and posttest require students to say out loud words or phrases to test their pronunciation, rather the tasks ask the students to use their current pronunciation knowledge to choose the correct answer Although the pretest and posttest aim at testing students’ pronunciation competency, it would be more reliable if students were requested to produce something orally Finally, the lack of inter raters is another factor that does not guarantee a high level the reliability and validity of the study.

R ECOMMENDATIONS FOR FURTHER STUDY

While this study has provided valuable insights into the impact of SpeechAce on developing pronunciation competency, its limitations presented above have called for a number of suggestions for further studies First of all, more studies with a larger sample size should be done on the effectiveness of using Automatic Speech Recognition technology in teaching and learning pronunciation Moreover, the pronunciation pretest and posttest if applied need to be more thoroughly designed to test all components of English pronunciation, especially, those tests must require test takers to produce words, phrases, or sentences orally to accurately gauge their level of pronunciation competency Also, two inter raters must also be involved in grading the pretest and posttest to maintain the objectiveness of the results

There are several avenues for further research that could enhance our understanding and contribute to the broader field of language education Researchers should also investigate the most effective pedagogical strategies for integrating ASR into language instruction Explore how teachers can optimally incorporate ASR into different teaching methods and curricula to maximize pronunciation learning outcomes We may also take diverse learners’ population into consideration by exploring how ASR caters to diverse learner populations, including individuals from various linguistic backgrounds, ages, and proficiency levels and assess whether the tool is equally effective for learners with different language learning needs Further studies might aim at finding out how the incorporation of ASR activities influences other language skills, such as listening comprehension and speaking fluency and investigate whether enhanced pronunciation contributes to broader language proficiency

Aiello, J., & Mongibello, A (2019) Supporting EFL Learners with a Virtual

Environment: A Focus on L2 Pronunciation Journal of e-Learning and Knowledge Society, 95-108

Brazil, D (1997) The Communicative Value of Intonation in English Book

Bryman, A., & Cramer, D (1990) Quantitative Data Analysis for Social

Celce-Murcia, M., Brinton, D., & Goodwin, J (2010) Teaching Pronunciation: A

Reference for Teachers of English to Speakers of Other Languages Cambridge: Cambridge University Press

Chapelle, C (2001) Computer Applications in Second Language Acquisition

Cheung, I., Gupta, A., & El-Shimi, A (2021, September 12) Retrieved from speechace.com: https://www.speechace.com

Cohen, L., Manion, L., & Morrison, K (2018) Research Methods in Education

Council, B (n.d.) Retrieved from ietls.org: https://www.ielts.org/-

/media/pdfs/speaking-band-descriptors.ashx?la=en-us

Couper, G (2015) Applying Theories of Language and Learning to Teaching

Pronunciation In M Reed, & J Levis, The Handbook of English Pronunciation (pp 413-432) New York: Wiley Blackwell

Creswell, J (2014) Research Design: Qualitative, Quantitative and Mixed

Creswell, J W (2012) Educational Research - Planning, Conducting, and

Evaluating Quantitative and Qualitative Research Boston: Pearson

Creswell, J., & Plano Clark, V (2017) Designing and Conducting Mixed Methods

Cunningham, U (2013) Teachability and Learnability of English Pronunciation

Features for Vietnamese Speaking Learners In E Waniek-Klimczak, & L

R Shockey, Teaching and Researching English Accents in Native and Non- native Speakers (pp 3-15) New York: Springer

Derwing, T., & Munro, M (2005) Second Language Accent and Pronunciation

Teaching: A Research-Based Approach TESOL Quarterly, 39(3), 379-397 Derwing, T., & Munro, M (2015) Pronunciation Fundamentals Evidence-based perspectives for L2 teaching and research John Benjamins

Derwing, T., Diepenbroek, L., & Foote, J (2012) How Well do General-Skills

ESL Textbooks Address Pronunciation? TESL Canada Journal, 22-44 Ellis, R (2006) Researching the effects of form-focused instruction on L2 acquisition AILA Review,, 19, 18-24

Ellis, R (2009) Implicit and explicit learning, knowledge and instruction In R

Ellis, S Loewen, R Erlam, C Elder, J Philp, & H Reinders, Implicit and explicit knowledge in second language learning, testing and teaching (pp 3-25) Bristol: Multilingual Matters

Foote, J., Trofimovich, P., Collins, L., & Urzúa, F (2013) Pronunciation teaching practices in communicative second language classes The Language Learning Journal , 181-196

Gilbert, J (2008) Teaching Pronunciation: Using the Prosody Pyramid

Gilbert, J B (2014) Clear Speech: Pronunciation and Listening Comprehension in North American English London: Cambridge University Press

Gorjian, B., Hayati, A., & Pourkhoni, P (2013) Using Praat Software in Teaching

Greene, J (2007) Mixed Methods in Social Inquiry San Francisco: Jossey-Bass Hancock, M (2003) English Pronunciation in Use Cambridge: Cambridge

Harmer, J (2016) Teaching Pronunciation In The Practice of English Language

Jenkins, J (2000) The Phonology of English as an International Language

Kelly, G (2000) How to Teach Pronunciation Longman

Kim, J., & Han, Z (2007) Recasts in communicative EFL classes: Do teacher intent and learner interpretation overlap In A Mackey, Conversational interaction in second language acquisition: A collection of empirical studies (pp 269-297) Oxford: Oxford University Press

Lee, J., Jang, J., & Plonsky, L (2015) Applied Linguistics The Effectiveness of

Second Language Pronunciation Instruction: A Meta-Analysis, 345-366 Levis, J (2005) Changing Contexts and Shifting Paradigms in Pronunciation

Levis, J (2007) Computer technology in teaching and researching pronunciation

Annual Review of Applied Linguistics, 184-202

Lidia, A M (2021) Overcoming segmental difficulties in English pronunciation in Spanish 3-ESO bilingual students through the use of SpeechAce Valladolid: Universidad de Valladolid

Long, M (2015) Second Language Acquisition and Task-Based Language

Lyster, R., Saito, K., & Sato, M (2013) Oral corrective feedback in second language classrooms Language Teaching, 46, 1-40

Morley, J (1991) The Pronunciation Component in Teaching English to Speakers of Other Languages TESOL Quarterly, 25(3), 481-520

Moxon, S (2021) Exploring the Effects of Automated Pronunciation Evaluation on L2 Students in Thailand IAFOR Journal of Education: Language Learning in Education, 41-56

Munro, M., Tracey, D., & Morton, S (2006) Studies in Second Language

Acquisition The Mutual Intelligibility of L2 Speech, 111-131

Newton, J (2018) Pronunciation and Speaking In O Kang, R Thomson, & J

Murphy, The Routledge Handbook of Contemporary English Pronunciation (pp 337-351) New York: Routledge

Nguyen Tan Loc; Jonathan Newton (2020) Pronunciation teaching in Tertiary

EFL classes: Vietnamese teachers' Beliefs and Practices 1-20

Patton, M Q (1980) Qualitative Evaluation Methods Beverly Hills: CA: Sage Pennington, M (1996) The Power of CALL Athelstan

Pennington, M (1999) Computer-Aided Pronunciation Pedagogy: Promise,

Limitations, Directions Computer Assisted Language learning, 427-440 Pennington, M., & Rogerson-Revell, P (2019) Intelligibility, Comprehensibility, and Interpretability In English Pronunciation Teaching and Research (pp 43-47) London: Palgrave Macmillan

Plonsky, L., & Mills, S (2006) An exploratory study of differing perceptions of error correction between a teacher and students: Bridging the gap Applied Language Learning, 55-74

Przedlacka, J (2018) An Overview of Phonetics for Lanugage Teachers In O

Kang, R Thomson, & J Murphy, The Routledge Handbook of Contemporary English Pronunciation (p 48) New York: Routledge Roach, P (2009) English Phonetics and Phonology: A practical course

Saito, K., & Lyster, R (2012) Effects of Form-Focused Instruction and Corrective

Feedback on L2 Pronunciation Development of /ɹ/ by Japanese Learners of English Language Learning, 62(2), 595-633

Samad, I., & Ismail (2020) ELSA Speak Application as a Supporting Media in

Enhancing Maspul Journal of English Studies, 1-7

Shadish, W., Cook, T., & Campbell, D (2002) Experimental and Quasi-

Experimental Designs for Generalized Causal Inference Houghton Mifflin Smith, Larry E.; Nelson, Cecil L.; (1985) World Englishes International intelligibility of English: directions and resources , 333-342

Spauling, S., & Breazeal, C (2019) Pronunciation-based Child-Robot Games

Interactions to Promote Literacy Skills 554-555

Suong, N (2016, November 16) zingnews.vn Retrieved from https://zingnews.vn/bo-truong-gd-dt-de-an-ngoai-ngu-9400-ty-khong-dat- muc-tieu-post698197.html

Tam, H (2005) Common Pronunciation Problems of Vietnamese learners of

English VNU Journal of Science, Foreign Languages, 35-46

Tuckman, B W (1972) Conducting Educational Research New York: Harcourt

Underhill, A (2005) Sound Foundations: Learning and Teaching Pronunciation

Vu, H Y (2016) Exploring English Pronunciation Teaching in Vietnam: Time for a New Approach?

A In each line, identify the word that has a different first consonant sound Example: friend priest physical philosophy

B In each line, identify the word that has a different final consonant sound

C In each line, identify the word that has a different number of syllables from the others

Example: lengths if table on

D In each line, identify the word that has a different stress pattern from the others

Example: picture nature capture mature

E Listen to the recording and decide which order they appear in

F Listen and fill in each gap with one or more words

2 (a) just got a present (b) _ my father

3 I think they must have gone (a) library

4 I’d like a glass or two (a) water

5 They thanked me (a) helping (b) find the money

G Write, in normal script, the name of these cities

A Circle the word with a different vowel sound

Example hot gold gone swan

B Add the consonant to the word to make another word

C Listen and circle the word you hear

1 Have you got a pan/ pin/ pen I could borrow?

2 We should clean the cut/ cat/ cot first

3 You won’t be able to feel/ fill/ fail this

4 I think that’s the west/ worst/ waist

5 The cot/ coat/ court’s too small

6 I don’t think it’s far/ fur/ fair, you know

7 What time did the woman/ women arrive?

9 I used to have a bet/ vet/ pet

10 I got a good price/ prize for it

11 They didn’t suit/ shoot him

12 I think it’s the code/ coat/ coach

13 That’s a fine/ wine/ vine colour

14 His back/ bag was broken

15 You can smell it in the air/ hair

16 You’ll have to watch/ wash the baby

18 I’ll collect/ correct/ connect it tomorrow

D Listen and circle the word you hear

1 The glass/ gas is green

2 The tooth/ truth is out

4 They need/ needed more time

5 I think they want/ wanted to talk

6 There are thirty/ thirteen people in my class

8 The books are/ were cheap

9 I think there are some pears and/ or grapes

10 What does/ did she say?

12 We need more sport/ support

13 It’s all in the past/ pasta now

14 Our guest/ guests came late

E Which word has a different number of syllables from the others? Circle it

Example snakes sheep foxes cats

F All the words or expressions in each group have the same number of syllables Circle the one with stress in a different place

Example October November December January

3 He told me I like it He finished Close the door

4 Go to bed! Don’t worry! What’s the time? Fish and chips

7 playground shoe shop first class handbag

G Listen Are the two expressions pronounced exactly the same on the recording, or is there a difference? Write S for same or D for different

Some of each summer beach …D…

1 Some of you Summer view

5 Burnt a cake Burnt the cake

7 Want to talk Wanted to talk

8 I’ve locked it I blocked it

10 A bitter fruit A bit of fruit

Questionnaire on teaching and learning English pronunciation with SpeechAce

This questionnaire seeks to collect your personal opinions about learning pronunciation with SpeechAce Your specific answers will help us improve our methods in teaching pronunciation in technology era Please take a few minutes to complete the questionnaire and rest assured that your information will be used for this research only Thank you very much for your cooperation

Instruction: Read the statement and tick (P) the option that you think is the most correct with you Give an answer for open questions

Ex: I think learning English is very easy ¨Strongly disagree ¨Disagree þNeutral ¨Agree ¨Strongly agree

Question Strongly disagree Disagree Neutral Agree Strongly agree

1 It is important for me to have good English pronunciation

2 I think learning pronunciation is interesting

3 Students should aim for native English pronunciation

4 The best way to perfect your pronunciation is by speaking to native speakers

5 I mainly learn pronunciation from textbooks

6 It is possible to achieve native like accent

7 Which of the following are the most useful in improving your pronunciation? (choose all that apply)

Know how to read phonemic transcription

Part 2: Difficulties in learning pronunciation

Question Strongly disagree Disagree Neutral Agree Strongly agree

8 I don’t know how to learn pronunciation by myself

10 I cannot find native speakers to practice my pronunciation

11 I find it hard to link the sounds together

12 I often miss the ending sound of a word

13 Teachers spend little time in teaching pronunciation

14 Learning pronunciation from different teachers’ accents confuses me

Question Strongly disagree Disagree Neutral Agree Strongly agree

15 SpeechAce is easy to use

16 My pronunciation has been improved after practicing with SpeechAce

17 I can read phonemic transcriptions after practicing with SpeechAce

18 I can pronounce ending sounds correctly by practicing with SpeechAce

19 I think SpeechAce can help me pronounce like a native speaker

20 I can learn pronunciation effectively on my own with SpeechAce

21 I know which part I pronounce incorrectly with detailed analysis of SpeechAce

22 I can imitate the reference pronunciation from

23 Do you like learning pronunciation with SpeechAce? Why/ why not?