ELT Echo : The Journal of English Language Teaching in Foreign Language Context (2022), Vol 7(2)

DOI: 10.24235/eltecho.v7i2.12503
Published by English Language Teaching Department, IAIN Syekh Nurjati Cirebon, Indonesia. p-ISSN: 2579-8170, e-ISSN: 2549-5089

ELT Echo : The Journal of English Language Teaching in Foreign

Language Context
journal homepage: https://syekhnurjati.ac.id/jurnal/index.php/eltecho

AI-POWERED APPS TO ENHANCE NOVICE NEWSREADERS’ ENGLISH

PRONUNCIATION

Ana Humardhiana
Department of Islamic Broadcasting and Communication, Faculty of Da’wah and Islamic Communication, IAIN Syekh Nurjati Cirebon,
Indonesia

*
Corresponding author: Perjuangan Street, Kesambi, Cirebon City, West Java, 45132, Indonesia. E-mail address: anahumardhiana@gmail.com
article i nfo abstract
Article history: Among the goals that Communication students, who are ESP learners, desire to achieve is
Received: 2 December 2022 to become professional newsreaders who can deliver the news in both Indonesian and
Received in revised form: 28 English. These novice newsreaders with little to no knowledge about news delivery in
December 2022 English often face problems with English pronunciation. This study aims to help forty-one
Accepted: 28 December 2022 novice newsreaders, who are students of the Islamic Broadcasting and Communication
Available online: 31 december 2022 Department, with their English pronunciation using AI-powered apps and to reveal their
Keywords: perceptions of the apps. The students were involved in one cycle of Classroom Action
AI Research (CAR), where they had to make a video of them reading English news before
Application using the apps as the pre-test and another video after the lecturer taught them how to use
Newsreader the apps as the post-test. The researcher distributed a questionnaire via Google Form after
Pronunciation the post-test to complete the data. The results show that AI-powered apps can enhance
ESP novice newsreaders’ English pronunciation, especially in the aspects of Accuracy, Key
Words, Chunking and Pausing, Intonation, as well as Sounds and Vocal Features; yet,
Intonation becomes the lowest improved aspect. Also, the respondents believed that the
apps help them with their future job as newsreaders, are easy to use, and give instant
feedbacks, which are necessary for independent learning and suitable for ESP learners. The
ELSA app rose to the top of the list of the respondents’ favorites. Finally, it is recommended
that future researchers carry out comparable studies that focus on one or two particular
English sounds evaluated across multiple cycles.

INTRODUCTION
The rapid development of technology has embarked on creating human-like intelligence
applied in any machine people often use in everyday life. From the automatic spelling
correction in Microsoft Word to Siri-infused home appliances, artificial intelligence (AI) has
significantly helped humans. It is an impactful innovation that characterizes The 4th Industrial
Revolution. According to McCarthy (2007), AI refers to the science and engineering of creating
intelligent machines, in which intelligence refers to the ability to solve problems
computationally to achieve goals. Russell (2010) states that AI is the art of constructing
computers capable of thinking and acting like humans or thinking and acting suitably. Al-
Shawabkah (2017, p. 23) adds that AI is described as the abilities transferred to computers to
enable many performance systems to be intelligent and to resemble humans in their behavior.
With these qualities of humans, AI is utilized to help people solve their problems and get
everyday work done with ease. One practical example of how people can use AI is via a
smartphone (Goksel & Bozkurt, 2019). The AI-powered applications in a smartphone help

214
people drive to a desired place with the assistance of GPS, solve algorithmic trading problems,
check and correct spelling errors in texts, and many others. Without question, AI is changing
the way people live, work, and learn (Manns, 2017).
In the education field, AI is a current trend to employ, especially for English learning
and teaching (ELT). AI-powered applications have hugely contributed to the teaching and
learning process of English. Haryadi and Aprianoto (2020) conducted a study to find out
whether the integration of an AI-based application called the English Pronunciation app can
increase students’ participation and self-learning in pronunciation classes at Mandalika
University of Education. The research results indicate that the integration of this app in teaching
pronunciation increased the students’ participation in terms of engagement, attitude, and
conduct. The app also brought a positive effect on the establishment of independent learning
for a significant number of students (Haryadi & Aprianoto, 2020). In the same year, Abbas and
Fathira (2020) conducted a further study to improve students’ pronunciation in pronouncing the
ending -ed by implementing an Android application and to find the factors influencing the
improvement. The results show that there was an improvement that can be learned from the
increase in the students’ ability to pronounce the ending –ed from the level of the “fairly good”
category to the “good” one. Meanwhile, the factors influencing the students’ ability to
pronounce the ending –ed are the students often practiced and listened to the Android
application either online or offline to obtain the understanding and the information on how to
pronounce the exact words of the ending –ed, which sounds /t/, /d/, and /id/ (Abbas & Fathira,
2020).
In 2021, Suciati conducted another study on students who lived in remote areas when
the COVID Pandemic hit to find out the students’ favorite AI-based speaking applications, the
underlying reasons, and the weaknesses of using AI-based speaking applications. Based on the
study, the favorite AI-based speaking apps are Cake, Talk, Elsa, and Speak English. The
students' reasons for choosing them are because those speaking apps were free, easy to access,
flexible to be used everywhere and every time, able to be used as good alternative speaking
partners, and able to give evaluation or assessment. Whereas, the weaknesses of the AI-based
speaking apps are limited to the topics and conversations served by the applications, so they
cannot be elaborated more. Unfortunately, the speaking applications with good features are not
freely accessed and need more space in the gadget (Suciati, 2021). Focusing on another skill to
research, Al-mawaly and AL-Jamal (2022) conducted a study aiming to investigate the effect
of AI on Jordanian EFL sixth-grade students' listening comprehension and their attitudes
towards it. The research results indicate that the experimental group with the help of AI
significantly scored higher than the control group without AI in the three and overall levels of
listening comprehension. The findings suggest that using AI effectively enhances EFL learners'
listening skill.
According to previous studies, AI-powered apps are proven to have a great impact on
students of English and/or students who learn English for general purposes. But not many
studies focus on the use of AI-powered apps for students who learn English for specific
purposes (ESP), such as for their jobs. This study is distinct in a way that it tries to help novice
newsreaders, who learn ESP, by applying AI-powered apps to improve their pronunciation for
better English news delivery. The novice newsreaders in this study are students of Islamic
Broadcasting and Communication, IAIN Syekh Nurjati Cirebon, who have some knowledge
about Indonesian news delivery but not enough knowledge about English news delivery.
Similar to Gilakjani’s study (2018), these students have problems with their pronunciation since
pronunciation is one of the aspects that are least likely to be taught by the lecturers in ESP
classes at this campus. In addition, these students do not have adequate exposure to English

215
since they do not live in an English environment where they can practice their speaking skill,
especially pronunciation. It is unfortunate since novice newsreaders' success in delivering the
news in English depends on their pronunciation, which they are still weak in. Thus, this research
attempts to help enhance novice newsreaders’ English pronunciation and reveal their
perceptions of the apps.

METHOD
The researcher applied the Classroom Action Research method in this research.
According to Ebutt in Hopkins (2014), Classroom Action Research (CAR) is the effort to
enhance teaching and learning through a series of practical actions and to reflect on the
outcomes of those actions. Kemmis and McTaggart (1998) define action research as an action
that is taken to inquire about one's own self-reflection and improve one's instruction by
evaluating one's own practice. Sagor (2010) sees action research as a process of inquiry
conducted by the practitioner and for those who take the action. Krathwohl (1993) adds that
action research is done by practitioners to improve practice. McNiff (2013) clarifies that the
researcher does his or her research in action research. In other words, the practitioner does the
research himself or herself to make some improvement on a specific inquiry. In this study, the
researcher conducted CAR in one of the classes she teaches to improve the students’ English
pronunciation by using AI-powered applications. Forty-one male and female students of the
Islamic Broadcasting and Communication Department, Faculty of Da’wah and Islamic
Communication, IAIN Syekh Nurjati Cirebon, were purposively selected as the participants of
this study. These students were in Semester 5 where they were supposed to have some practice
in delivering the news in Indonesian and in English. They were treated like novice newsreaders
who still had no experience in news delivery.
In this study, the researcher used the basic model of CAR developed by Kemmis and
McTaggart (2013), which has four stages in a cycle; CAR allows the implementation of a series
of cycles to solve the problem. Below is the figure of a cycle by Kemmis and McTaggart (2013).

Planning

Reflection Action

Observation

Figure 1. Four Stages of A CAR Cycle by Kemmis and McTaggart (2013)

The initial stage that the researcher did was Planning. In this stage, the researcher designed the
lesson plan, materials, technique, pre-test, post-test, interview questions, and pronunciation
accuracy checklist for the assessment. The researcher also selected some AI-powered
applications to recommend to the students. The next stage is Action, where the researcher
conducted all the planned activities. In this stage, the researcher instructed the students to make
a video where they read a piece of English news as the pre-test. After that, in the following

216
meeting, the researcher taught the students how to use AI-powered apps to check their
pronunciation. Then, they were instructed to make another video where they read the same
English news as the post-test. This step was done to see how their pronunciation accuracy had
improved. A pronunciation accuracy checklist adapted from the Pronunciation Intensive
Academic Program e-booklet by the University of Technology, Sydney (2016) was employed
to gain the quantitative data. However, the researcher needed to adjust the aspects to assess,
which are Accuracy, Key Words, Chunking and Pausing, Intonation, and Vocal Features. The
researcher used the institute’s grading system to score the pre-test and the post-test.
Table 1. Grading System of IAIN Syekh Nurjati Cirebon
Score Grade Quality of Achievement
90-100 A Excellent
85-89 A-
Very Good
80-84 B+
75-79 B
Good
70-74 B-
65-69 C+
Satisfactory
60-64 C
50-59 D
Failure/No Credit
0-49 E

Also, a questionnaire via Google Form was carried out after the post-test to complete the data.
In the Observation stage, the researcher observed and monitored how the students used the AI-
powered apps. She also administered the assessment using the pronunciation accuracy
checklist. The last stage was Reflection, where the researcher reflected on how the research had
been carried out by learning from the analysed data. Since the learning progress was already
visible in one cycle, the researcher did not conduct multiple cycles.

FINDINGS AND DISCUSSION

Findings
The pre-test and the post-test were conducted in two different meetings, Week 12 and
Week 13. In Week 12, the students did the pre-test (P1), where they took a video of themselves
reading news from ABC7 Channel about an earthquake hitting Cianjur that drew international
attention. The news was taken from https://www.youtube.com/watch?v=FDs-BhvwAmI. Then,
the students were taught how to use AI-powered apps to enhance their English pronunciation.
At the end of the class, they were instructed to make a video where they read the same English
news within a week of submission, as the post-test (P2). Hence, they had seven days to train
their pronunciation using the apps. This step was conducted to know how these apps affected
their pronunciation. The students were given the freedom to choose which app to use. So, one
student might have a different app from another. Forty-one students were involved in this
research and submitted their work. The researcher assessed their pronunciation by employing a
pronunciation accuracy checklist adapted from the Pronunciation Intensive Academic Program
e-booklet by the University of Technology, Sydney (2016) which measures pronunciation
aspects that can be used for news delivery, namely Accuracy (how accurate the words are
pronounced based on the standard), Key Words (whether the keywords stressed), Chunking and
Pausing (whether the information delivered divided into chunks or thought groups and whether
the speaker used pausing appropriately), Intonation (whether the speaker’s intonation indicate
finished and unfinished information and whether the pitch range was wide enough to make the
most important keywords easy to hear as well as to make the speaker sound interesting), and
Sounds and Vocal Features (whether the sounds and syllables linked together, as well as

217
whether the speed and the volume were just right). Below is the result of the assessed pre-test
(P1) and post-test (P2) from all forty-one students in one cycle.
Table 2. Results of the Pre-Test (P1) and Post-Test (P2)
Aspects P1 Average P2 Average Improvement (%)
Accuracy 58 80 37.93%
Key Words 45 62 37.78%
Chunking and Pausing 48 79 64.58%
Intonation 57 73 28.07%
Sounds and Vocal Features 59 76 28.81%
Mean 53.4 74 38.58%

The first column of Table 2 displays the aspects of pronunciation assessed. The second
column shows the average P1 score of all 41 students from all aspects. In P1, the Accuracy
aspect’s score is 58, which is Failure. The Key Word aspect in P1 average score hits not more
than 45 (Failure). The Chunking and Pausing aspect is slightly no different by gaining only
three more points or 48 in total, still in the Failure quality. The Intonation aspect in P1 average
score gets 57 points, which is, again, still in the Failure quality. The highest score of all aspects
in P1, which is 59 points, is gained by the Sounds and Vocal Features. However, it still falls in
the Failure quality. From all five aspects, the mean of the P1 average score is 53.4, which falls
under the institute’s grade category of D (Failure). The third column of Table 2 shows the
average P2 score from the five aspects. The Accuracy aspect in the P2 score hits 80 points,
which increases 37.93% from P1. This score falls in the quality of Very Good. The Key Word
aspect in the P2 score is increased by 37.78% or 23 points, from 45 in P1 to 62 in P2, which
falls in the quality of Satisfactory. The Chunking and Pausing aspect has the highest
improvement percentage of 64.58%, from 48 to 79 in P2, which falls in the quality of Good.
While in the aspect of Intonation, there is an increase of 28.07%, from 57 points to 73 points in
P2, which falls in the quality of Good, and there is also an increase of 28.81% in the Sound and
Vocal Feature aspect in P2, from 59 to 76 points, which falls in the quality of Good. Overall,
the students’ pronunciation improved by 38.58%, from 53.4 points in P1 to 74 points in P2,
which meets the quality of Good. Based on the description, we can see that the most improved
aspect in one cycle is Chunking and Pausing. The respondents were seen to be able to divide
information into chunks or thought groups and use pausing appropriately. Meanwhile, the least
improved aspect is Intonation. The respondents could still not use Intonation (whether the
speaker’s intonation indicates finished and unfinished information and whether the pitch range
was wide enough to make the most important keywords easy to hear as well as to make the
speaker sound interesting).
Besides conducting a pre-test and a post-test for the students, the researcher distributed
a questionnaire to the students to complete the data. The result of the questionnaire is displayed
in a figure and tables below.

218
 App Users
ELSA

Speakometer
2%
2% Duolingo
22% 34% Cakap

English Pronunciation by
KEPHAM
15% English Pronunciation by
7% az-20 Apps
3% 15% English Pronunciation by
Awabe
Others

Figure 1. Most Used AI-Powered Pronunciation Apps

Figure 1 shows the AI-powered pronunciation apps most used by the students. Based on the
result, ELSA became the most favorite app to use by 34% users of the forty-one students. The
runner-up place is seated by English Pronunciation by az-20 Apps with 22% users. Duolingo
and English Pronunciation by KEPHAM got the same percentage of users, which is 15%. The
next app is Speakometer which drew the interest of 7% of students. Cakap is preferred by 3%
of the students. The least favored app is English Pronunciation by Awabe with 2% of users.
The rest students, 2% in percentage, utilized the AI in Google Translate machine. Based on the
figure, we can see that ELSA became the most favorite application to use to improve the
students’ pronunciation. Most of the respondents already knew what ELSA was. They only had
not had the chance to install and use it for a visible purpose.
The questionnaire also asked whether the app chosen by the students helped them
enhance their English pronunciation. The result is shown in Table 3.
Table 3. This App Helps Me Enhance English Pronunciation
Statement Number of Respondents Percentage
Strongly Agree 13 31.7%
Agree 21 51.2%
Neutral 5 12.2%
Disagree 2 4.9%
Strongly Disagree - -

Table 3 shows that 31.7% of the respondents, which are 13 students in the figure, strongly
agreed that the app they chose helped them improve their English pronunciation. There were
51.2% of the respondents, or 21 students agreed that the apps helped them with their English
pronunciation. As many as five students, or 12.2% of the respondents, had a neutral opinion
about this topic. While two students, or 4.9% of the respondents, disagreed that the apps helped
them improve their pronunciation. And zero students strongly disagreed that the apps could
enhance the users’ English pronunciation. From the table, we can see that the majority of the
students think that the apps they chose helped them enhance their English pronunciation.
The students were also asked to respond to a statement about whether the app they chose
gave accurate feedback. The result is displayed in Table 4 below.

219
 Table 4. This App Gives Accurate Feedback
Statement Number of Respondents Percentage
Strongly Agree 11 26.8%
Agree 20 48.8%
Neutral 7 17.1%
Disagree 2 4.9%
Strongly Disagree 1 2.4%

Table 3 reveals that eleven students, or 26.8% of the respondents, strongly agreed that the apps
gave accurate feedback. As many as 20 students, or 48.8% of the respondents, agreed with this
statement. Seven students, or 17.1% of respondents, had a neutral opinion of the statement. Two
students, or 4.9% of the respondents, disagreed that the apps gave accurate feedback.
Meanwhile, one student, which made up 2.4% of the respondents, strongly agreed with the
statement. From the description, we can say that the majority agreed that the apps gave accurate
feedback on their learning.
The next statement to respond to is whether the app they chose was easy to use. The
result is presented in Table 5 below.
Table 5. This App Is Easy To Use
Statement Number of Respondents Percentage
Strongly Agree 15 36.6%
Agree 19 46.3%
Neutral 6 14.6%
Disagree - -
Strongly Disagree 1 2.4%

Table 5 shows fifteen students, or 36.6% of the respondents, strongly agreed that the app they
chose was easy to operate or use. As many as 19 students, or 46.3% of the respondents, agreed
with the statement. It is found that not all respondents agreed with this statement because six
students, or 14.6% of the respondents, had a neutral opinion. Meanwhile, there were no
respondents who disagreed with the statement. Yet, one student, or 2.4% of the respondents,
strongly agreed that the app chosen was easy to use. From the description, we can say that the
majority agreed that the apps were easy to use.
The students were also asked to respond to a statement about whether the app they chose
helped them learn independently or autonomously. The result is displayed in Table 6 below.
Table 6. This App Helps Me Learn Independently
Statement Number of Respondents Percentage
Strongly Agree 18 43.9%
Agree 21 51.2%
Neutral 2 4.9%
Disagree - -
Strongly Disagree - -

Table 6 reveals that there were eighteen students, or 43.9% of the respondents, who strongly
agreed that the app they chose helped them learn independently. Twenty-one students, or 51.2%
of the respondents, agreed with this statement. There were only two students, or 4.9% of the
respondents, who had a neutral opinion about this statement, while there were no respondents
who disagreed and strongly disagreed with the statement. From the description, we can learn
that the majority agreed that the apps helped the respondents or users learn English
pronunciation independently.

220
 The last statement to respond to is about whether the app they chose helped them as a
newsreader in the future. The result is presented in the following table.
Table 7. This App Helps Me As A Newsreader In The Future
Statement Number of Respondents Percentage
Strongly Agree 14 34.1%
Agree 22 53.7%
Neutral 3 7.3%
Disagree 2 4.9%
Strongly Disagree - -

Table 7 shows fourteen students, or 34.1% of the respondents, strongly agreed that the apps
they chose helped them for their future job as newsreaders. There are twenty-two students, or
53.7% of the respondents, who agreed with the statement. While three students, or 7.3% of the
respondents, had a neutral opinion, and only two students, or 4.9% of the respondents, disagreed
with this statement. There was no single respondent who strongly disagreed with the statement.
From the description, we can learn that the majority agreed that the apps helped the respondents
with their future job as newsreaders.

Discussion
Based on the research findings, AI-powered apps are proven to be able to help novice
newsreaders enhance their English pronunciation. It is in line with Kholis (2021) who states
that ELSA Speak, an AI-based application, can increase students’ pronunciation skills. After
seven days of practice, the novice newsreaders made some improvement in all aspects of
pronunciation assessed, namely Accuracy, Key Words, Chunking and Pausing, Intonation, as
well as Sounds and Vocal Features. All of the aspects had been in the quality of Failure in the
beginning. Yet, after the novice newsreaders practiced their pronunciation using AI-powered
apps, they achieved varied qualities, from satisfactory to very good at the end of one cycle.
Practicing English, pronunciation in this case, was never done previously even though most of
them knew that the apps, especially ELSA as the most favorite and well-known app, could help
them improve their English, specifically pronunciation. They did not have the chance to install
and use them for something relating to their major. It becomes the role of ESP lecturers to
facilitate and teach them how to benefit from the apps.
The aspects of English pronunciation assessed are very significant for English news
delivery. Based on the findings, all aspects showed some progress. However, the Intonation
aspect made the lowest progress by only 28.07% improvement, even though discourse
intonation plays a significant role in the successful achievement of professional communicative
goals (Nihalani & Lin, 1998). It becomes a note for lecturers to implement a new technique
regarding teaching how to make students’ intonation indicate finished and unfinished
information and how to make the pitch range wide enough so that the most important keywords
can be easy to hear as well as the speaker can sound interesting, which is crucial for English
news delivery. The lecturers can provide various English news scripts and select some words
and phrases potentially mispronounced and given the wrong intonation for the students to
practice using AI-powered apps, which can be implemented in the following cycles. For a more
measurable result, the lecturers can opt only one application to teach. However, this is
applicable only for classes with the same socio-economic background since not all students in
Indonesia can possess compatible smartphones. Even so, it is proven in this research that AI-
powered apps are easy to use and able to promote autonomous learning for students who learn
English for specific purposes (ESP) since they give instant feedback.

221
CONCLUSION
To sum up, AI-powered apps can enhance novice newsreaders’ English pronunciation,
especially in the aspects of Accuracy, Key Words, Chunking and Pausing, Intonation, as well
as Sounds and Vocal Features. From the five aspects of pronunciation improved, Intonation
becomes the aspect in which the novice newsreaders are weak. Thus, the ESP lecturers have to
implement a technique where the students can have more practice and drilling to make the
students’ intonation indicate finished and unfinished information, the pitch range wide enough
so that the most important keywords can be easy to hear, as well as the speaker can sound
interesting. The lecturers can provide various English news scripts and select some words and
phrases that are potentially mispronounced and given the wrong intonation for the students to
practice using AI-powered apps. The findings also show that the respondents believed the apps
help them with their future job as newsreaders, are easy to use, and give instant feedbacks,
which are necessary for independent learning. Among the apps, ELSA became the most favorite
app. Lastly, it is recommended that future researchers conduct similar research which
concentrates on one or two specific English sounds assessed in more than one cycle.

REFERENCES
Abbas, M. F. F., & Fathira, V. (2020). Implementation of Android application to solve the
students’ pronunciation of ending -ed. English Review: Journal of English Education, 8(2),
265-272. doi: 10.25134/erjee.v8i2.2648.
Al-mawaly, H. M., & AL-Jamal, D. A. (2022). The Effect Of Artificial Intelligence Application
On Jordanian Efl Sixth-Grade Students' Listening Comprehension And Their Attitudes
Towards It. Journal of Positive School Psychology, 8781-8791.
Al-Shawabkah, A. A. (2017). The role of AI applications (Expert Systems) in making
administrative decisions in the General Saudi Banks, Ta’if Governorate. Ta’if University
Magazine on Humanities, 49(15), 13−59.
Gilakjani, P. (2018). Teaching Pronunciation of English with Computer Technology: A
Qualitative Study. International Journal of Research in English Education. DOI:
10.29252/ijree.3.2.94
Goksel, N., & Bozkurt, A. (2019). Artificial intelligence in education: Current insights and
future perspectives. In Handbook of Research on Learning in the Age of
Transhumanism (pp. 224-236). IGI Global.
Haryadi, S., & Aprianoto. (2020). Integrating “English pronunciation” app into pronunciation
teaching: How it affects students’ participation and learning. JOLLT Journal of Languages
and Language Teaching, 8(2), 202-212.
Hopkins, D. (2014). A teacher's guide to classroom research. McGraw-Hill Education (UK).
Kemmis, S., & McTaggart, R. (1998). The nature of action research. The Action Research
Planner. Victoria: Deakin University.
Kemmis, S., McTaggart, R., & Nixon, R. (2013). The action research planner: Doing critical
participatory action research. Springer Science & Business Media.
Kholis, A. (2021). Elsa speak app: automatic speech recognition (ASR) for supplementing
English pronunciation skills. Pedagogy: Journal of English Language Teaching, 9(1), 01-
14.
Krathwohl, D. R. (1993). Methods of educational and social science research: An integrated
approach. Longman/Addison Wesley Longman.
McNiff, J. (2013). Action research: Principles and practice. Routledge.
Manns, U. N. E. S. C. O. (2017). Artificial Intelligence: Opportunities, threats and the future of
learning.

222
McCarthy, J. (2007). What is artificial intelligence.
Nihalani, P., & Lin, T. P. (1998). Intonation patterns in news broadcasts. World
Englishes, 17(1), 15-29.
Russell, S. J. (2010). Artificial intelligence a modern approach. Pearson Education, Inc..
Sagor, R. (2010). The action research guidebook: A four-stage process for educators and
school teams. Corwin Press.
Suciati, S. (2021, June). Favourite Artificial Intelligence-based speaking applications for
autonomous learning. In UNNES-TEFLIN National Seminar (Vol. 4, No. 1, pp. 205-214).
University Of Technology, S. (2016). Pronunciation Intensive academic program. 1–76.
www.helps.uts.edu.au

223