The particular dialect or language that a person chooses to use on any occasion is called a code 417 Copyright © 2024 The Author IDEAS is licensed under CC-BY-SA 4.0 License Issued by English study program of IAIN Palopo IDEAS Journal of Language Teaching and Learning, Linguistics and Literature ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) Volume 12, Number 1, June 2024 pp. 417 - 440 Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) Namira Fasya Khailifa1 , Pirman Ginting2 namirafasya1305@gmail.com pirmanginting@umsu.ac.id 12 Pendidikan Bahasa Inggris, Universitas Muhammadiyah Sumatera Utara, Medan Received: 2024-03-06 Accepted: 2024-04-01 DOI: 10.24256/ideas. v12i1.4908 Abstract AI (Artificial intelligence) is a design of technology that has been significantly useful for human in all aspects, particularly in education. In education systems, AI has become essential in providing the needs of students especially in english speaking skills. The development of AI (Artificial Intelligence) technology has enhanced the cognition and practices of EFL students in applying english speaking skills. This current study attempts to unleash the extent to which the speaking skills obtained by the EFL students with the use of AI. 15 EFL students in SMA N 10 Medan participated in this study. The research design employed mix method included qualitative and quantitative method. Pre-test and post-test was employed as an instrument of collecting data. The technique of analyzing data involved descriptive statistics and paired sample t-test were conducted to achieve the quantitative analysis. Meanwhile, thematic analysis was conducted to acquire the qualitative analysis. The result revealed that the majority of participants believed their cognition and practice in english speaking skills have improved with the assistance of AI. The findings also revealed that AI considered as fun, easy and practical alternative to practice english speaking proficiency. Therefore, it is suggested to all students could access AI as a tool that supports continuity of learning. Keywords: Artificial Intelligence, EFL Students, Speaking Skills http://u.lipi.go.id/1457703302 mailto:namirafasya1305@gmail.com mailto:pirmanginting@umsu.ac.id Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 418 Introduction In the past few years, technology has significantly become an alternative to give an assistance in our daily lives, especially in education. In education systems, most of the students were using technology in order to help find learning materials or completing assignments both at school and at home. One of the most technology that EFL students have used recently is AI (Artificial Intelligence). The use of AI (Artificial Intelligence) plays an important and bigger role in education (Khosravi et al., 2022). Artificial Intelligence (AI) is becoming a more significant part of everyday life, so it’s critical to comprehend its processes and applications. Consequently, the initial phases of education are incorporating principles of artificial intelligence (Kim & Kwon, 2023). When discussing artificial intelligence (AI), it is worth nothing that this technology has found extensive application in the realm of education, namely in the context of English as a foreign language (EFL) instruction. One of the challenges encountered by English as a Foreign Language (EFL) learners pertains to oral communication skills or speaking skills. Speaking has taken a very vital part in english language learning. Speaking is a skill that considerably increases the communication knowledge of EFL learners to the outside world. Teaching speaking is a useful talent that is typically thought to be challenging to acquire since it requires a lot of practice and perseverance to reach a high level of competency (Masuram & Sripada, 2020). Nowadays, speaking was mostly taught as a covert curriculum that was not included in official schooling (Hulsbergen et al., 2022). We might probably frequently witness some learnears using their speaking skills in the wrong way. For example, how to say greetings and goodbye, how to ask and offer for help, how to ask or offer for suggestions, how to pronounce a good vocabulary, how to deliver a news, etc. In fact, it is evident that students have not effectively utilized technology, such as social media, despite the significant role these platforms play in enhancing speaking proficiency. As we commonly understood, while considering the enhancement of speaking abilities in English as a Foreign Language (EFL) learners, AI (Artificial Intelligence) is a technological system that was designed to guide our works more neatly. Artificial intelligence (AI) has been employed for numerous years to assist in diverse tasks such as voice recognition, handwriting analysis, image recognition, and natural language processing (Southworth et al., 2023). Students encounter numerous challenges in their pursuit of improving their speaking skills. The utilization of artificial intelligence (AI) in the realm of speaking skills is anticipated, as it holds promise for addressing many challenges associated with verbal IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 419 communication. Therefore, AI (Artificial Intelligence) has the competency to lead students to get a good speaking skill. Discovering the knowledge about artificial intelligence (AI) can enhance students' employability and provide them with valuable insights to prepare for emerging job opportunities and potential mentorship (Southworth et al., 2023). Recent studies have shown that AI has an impact to led the EFL learners to create a sophisticated intelligent real-time feedback in writing skills (Gayed et al., 2022). Moreover, AI has been successfully created a program called “AI Across the Curriculum” which aims to provide each undergraduate student a chance to interact with and learn about AI in the context of their field and more crucially, in an interdisciplinary approach is frequently more representative of the actual workplace (Southworth et al., 2023). Speaking through the use of AI as a means of communication might enable English as a Foreign Language (EFL) learners to effectively express information and comprehend the intended significance of their statements. The integration of artificial intelligence (AI) in language learning is anticipated to foster the engagement of English as a Foreign Language (EFL) learners in the acquisition of speaking abilities, encompassing both foundational and advanced proficiencies. With the aid of AI, EFL learners are able to encourage their ability and skills in speaking. AI (Artificial Intelligence) was invented to test the level of their speaking such as, fluency, tone, vocabulary, etc. However, there has been a little development on how AI works for this skill. In addition, limited study has been conducted on the utilization of Artificial Intelligence (AI) for enhancing speaking skills among English as a Foreign Language (EFL) learners. Regarding this issue, this study endeavors to explore the extent to which the speaking abilities and skills obtained by EFL learners and what cognition that EFL Students achieved when using AI in speaking skills. With the support of AI, EFL learners could maximize and optimize the performance of their speaking skills. Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 420 Method The researcher used mix method namely, quantitative and qualitative methods. In order to shed insight into the depth of students' comprehension and proficienc y in speaking skills, this study employs quantitative research methods. Quantitative research is a valuable approach that enables the analysis of variables in relation to one another (Creswell, 2014). In order to accurately assess students' speaking abilities, it is crucial to employ effective measurement techniques. On the other hand, researchers also employ qualitative research as a method to delve into and comprehend the significance that individuals or groups attribute to a social human problem (Creswell, 2012) in which this study will examine the process and implementation of AI, as well as the advantages and challenges encountered by EFL students. In order to uncover this information, researchers will conduct interviews with multiple English as a Foreign Language (EFL) students, focusing on the method, advantages, and challenges associated with utilizing artificial intelligence (AI) technology to enhance speaking skills. The study sample comprised a 11th-grade class, with participants ranging in age from 16 to 17 years old by using a simple random sampling method. Simple random sampling is a widely employed sampling technique in scientific research (Tajik & Golzar, 2022). To conduct this sampling, the researcher will select a population consisting of 15 pupils in the 11th grade class. The determination of the sample size for this study will be made by the researcher. Subsequently, the researcher will employ a random selection method to choose volunteers from the entire population. This research instrument was conducted with experiments or trials in the form of one group pre-test and post-test design. The one-group pre-test and post- test design were expected to be given before and after practicing English speaking through the designated artificial intelligence (AI) programs. A one-group pretest- posttest design is a quasi-experimental research design that involves measuring the same dependent variable in a set of subjects both before (pretest) and after (posttest) administering a treatment. According to (Zou et al., 2023), Moreover, these tests were conducted to determine whether there were any alterations in their proficiency in spoken English. This research employs Campbell and Stanley (1963) model as the data collecting technique. Campbell and Stanley state that quasi-experiments strive to achieve the same level of rigor as true experiments, while recognizing the constraints imposed by real-world conditions. IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 421 This procedure of the study consisted in three stages. The first stage, researcher instruct the participants on enhancing their speaking skills accurately and effectively, without using the technology of AI. In the second stage, the participants were required to volunteered in the one group pre-test and post-test. The students will obtain insight into their abilities and capabilities through the scores in speaking skills. There are three steps that participants should follow in the one group pre-test and post-test. Firstly, before starting the teaching process, a pre-test was administered to assess their competence level, fluency, pronunciation, vocabulary usage, and overall communication skills. Secondly, interventions are implemented. The interventions or treatments that researcher conducted was using AI. By utilizing the technology of AI, it can encompass a range of activities, including interactive speaking exercises, role-plays, conversations, presentations, language games, and feedback sessions. Finally, the researcher will utilize a post- test to assess the individuals' proficiency in spoken English. The identical evaluation criteria utilized in the initial examination are employed to measure any enhancements in their spoken communication abilities. In the last stage, students will be faced with several questions through interviews regarding the process, benefits, and difficulties in carrying out speaking skills using AI. The research's success will be determined by the oral proficiency rating categories outlined by Brown (2001). Score Aspects Grammar Vocabulary Comprehension Fluency Pronunciation Task 1 Grammar errors are common, yet a native speaker accustomed to interacting Speaking vocabulary is insufficient to convey anything beyond basic necessities. He can comprehend basic inquiries and assertions when presented with slow speech, repetition, or The fluency descripti on is not specified . Consult the remainin Common pronunciation errors can be understandable t o a native speaker accustomed to interacting with non-native Can inquire and respond to inquiries on subjects Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 422 with non- native speakers can still understand the speaker. paraphrasing due to his restricted language skills. g four language categorie s to determin e the implied degree of proficien cy. speakers trying to speak their language. he is well knowled geable about. Capable of meeting basic travel needs and minimu m civility expectat ions. Should possess the ability to order a basic meal, inquire about and provide basic directio ns, make transacti ons, and tell time. 2 Can accurately handle basic Possesses a verbal vocabulary I can comprehend the main idea of Can confiden tly The accent is understandable yet frequently Capable of meeting IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 423 constructio ns but needs more complete and confide nt control of grammar. that allows him to communicat e himself clearly, albeit with considerable use of indirect language. most discussions on non-technical topics. (e.g., subjects that do not necessitate specific expertise) navigate most social situation s, including introduct ions, casual chats about current events, jobs, family, and personal informat ion. flawed. regular social expectat ions and job responsi bilities; need assistan ce managin g any complex ities or challeng es. Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 424 3 Grammar control is excellent. Capable of speaking the language accurately enough to engage effectively in a wide range of formal and informal discussions on practical, social, and professional subjects. Proficient in the language with a broad vocabulary to engage effectively in a wide range of formal and informal discussions covering practical, social, and professional subjects. His vocabulary is extensive enough that he often needs help finding the right term. Understanding is thorough when communication is delivered at a standard pace. Can easily discuss specific areas of expertise . Rarely struggles to find words. Mistakes do not hinder comprehension and seldom bother native speakers. The accent may indicate a foreign origin. Capable of engagin g proficie ntly in formal and casual discussi ons covering practical , social, and professi onal subjects. IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 425 4 Proficient in using language accurately at all levels required for professional purposes. Grammatica l errors are infrequent. Can comprehend and engage in conversation s within his realm of knowledge using a precise and extensive vocabulary. Can comprehend any dialogue within the scope of his familiarity. Proficien t at utilizing the language at a professio nal level. Can engage in discussio ns within this field with a high level of fluency. Pronunciation errors are uncommon. It may not be mistake n for a native speaker, but it can answer well in unusual circumst ances. Proficie nt in informal interpre ting from and into other languag es. Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 426 5 Comparable to that of a well- educated native speaker. Educated native speakers readily accept speech at all levels, including vocabulary, idioms, colloquialism s, and cultural references. Comparable to the proficiency level of a well- educated native speaker. He is fluent in the language to the extent that his speech is complete ly accepted by educated native speakers . Equivalent to and widely acknowledged by well-educated native speakers. Fluent in spoken languag e at a level compara ble to a well- educate d native speaker. The focus of this study was analyzed based on the research questions. The data was obtained through a quantitative analysis. This study employs descriptive statistics as a technique to analyze the data of EFL learners in improving or understanding their speaking skills. Descriptive statistics are employed to succinctly present data by illustrating the connection between variables in a sample or population (Kaur et al., 2018). To explore the development of AI in speaking skills, several paired sample t-tests were conducted. Meanwhile, to acquire a thorough quantitative analysis the statistical software IBM SPSS19 was employed to calculate various statistical measures including the mean, median, standard deviation, significance, lower, and upper. On the other hand, the qualitative data was analyzed by using an interview. Interview data was recorded and transcribed. In order to gather qualitative data for this research, the method of thematic analysis was employed. Thematic analysis refers to the systematic approach of identifying and analyzing patterns or themes that emerge from qualitative data (Maguire & Delahaunt, 2017). The technique being discussed in this article involves the analysis of data to identify, analyze, and interpret qualitative data patterns. These patterns are derived from the results of interviews that have been conducted previously. This research employs Braun and Clarke (2006) data analysis technique. Braun and Clarke's thematic analysis method is a sequential procedure comprising six iterative steps: The process involves six steps: IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 427 (1) acquainting oneself with the data, (2) creating codes, (3) identifying themes, (4) examining the identified themes, (5) defining and labeling the themes, and (6) finding examples that represent the themes (Mihas, 2023) Results The Improvement of Practice English Speaking Skills Through the Use of AI (Artificial Intelligence) After administered all the steps above, the researcher formulated the findings. 15 participants have volunteered to participate in pre-tests and post-tests in practicing English speaking skills. The participants were students from second grade year in SMA N 10 Medan. The results from paired-sample t-tests to compare the mean scores of the pre-test and post-test confirmed participants improvement in applying different speaking skills. Table 1. The Results of Paired-Sampled t-Tests for Speaking Skills 95% CI M SD t Sig Lower Upper Grammar Pre-test 2.46 1.06 -11.000 .000 -1.752 -1.180 Post-test 3.93 798 Vocabulary Pre-test 3.13 833 -11.225 .000 -1.429 -970 Post-test 4.33 816 Comprehension Pre-test 3.00 845 -16.000 .000 -1.209 -923 Post-test 4.06 798 Fluency Pre-test 2.93 798 -10.693 .000 -1.680 -1.119 Post-test 4.33 723 Pronunciation Pre-test 3.06 883 -12.475 .000 -1.328 -938 Post-test 4.20 861 Task Pre-test 2.86 743 -12.475 .000 -1.328 -938 Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 428 As indicated in Table 1, the mean grammar score for speaking has increased. The scores increased from the pre-test (M= 2.46, SD= 1.06) to the post-test (M= 3.93, SD= 798). The increase showed statistical significance (p <.000) with a 95% confidence interval (CI) between -1.752 and -1.180. Similarly, the mean vocabulary score has risen. From the pre-test (M= 3.13, SD= 833) to the post-test (M= 4.33, SD= 816). The increase showed statistical significance (p <.000) with a 95% confidence interval (CI) between -1.429 and -970. The comprehension score significantly increased from the pre-test (M= 3.00, SD= 845) to the post-test (M= 4.06, SD= 798). The increase was statistically significant (p <.000) with a 95% confidence interval (CI) from -1.209 to -923. Additionally, there was a notable improvement in fluency from the pre-test (M= 2.93, SD= 798) to the post-test (M= 4.33, SD= 723). The increase showed statistical significance (p <.000) with a 95% confidence interval (CI) between -1.680 and -1.119. The pronunciation score significantly increased from the pre-test (M= 3.06, SD= 883) to the post-test (M= 4.20, SD= 861). The increase was statistically significant (p <.000) with a 95% confidence interval (CI) ranging from -1.328 to -938. Ultimately, the average task score also increased. The scores increased from the pre-test (M= 2.86, SD= 743) to the post-test (M= 4.00, SD= 755). The score increased dramatically with a 95% confidence interval (CI) ranging from -1.328 to -938. In conclusion, the average scores for various speaking skills showed a significant increase compared to the original scores. The scores for grammar, fluency, and pronunciation showed the most improvement. The findings showed that AI significantly enhanced participants' various speaking skills. Table 2. The Results of Descriptive Statistics for Speaking Skills N Minimum Maximum Mean Std. Deviation Pre Grammar 15 1.00 4.00 2.4667 1.06010 Post Grammar 15 3.00 5.00 3.9333 .79881 Pre Vocabulary 15 2.00 4.00 3.1333 .83381 Post Vocabulary 15 3.00 5.00 4.3333 .81650 Pre Comprehensi 15 2.00 4.00 3.0000 .84515 Post-test 4.00 755 IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 429 on Post Comprehensi on 15 3.00 5.00 4.0667 .79881 Pre Fluency 15 1.00 4.00 2.9333 .79881 Post Fluency 15 3.00 5.00 4.3333 .72375 Pre Pronunciatio n 15 2.00 4.00 3.0667 .88372 Post Pronunciatio n 15 3.00 5.00 4.2000 .86189 Pre Task 15 2.00 4.00 2.8667 .74322 Post Task 15 3.00 5.00 4.0000 .75593 Valid N (listwise) 15 According to the descriptive statistics presented in Table 2, there has been an increase in both the minimum and maximum grammar scores in speaking. There was an improvement from the pre-test scores (Min= 1.00, Max= 4.00) to the post-test scores (Min= 3.00, Max= 5.00). The range of vocabulary scores saw a significant increase. The scores increased from the pre-test (Min= 2.00, Max= 4.00) to the post-test (Min= 3.00, Max= 5.00). Likewise, the minimum and maximum comprehension scores have been raised. From the pre-test scores (Min= 2.00, Max= 4.00) to the post-test scores (Min= 3.00, Max= 5.00). The range of scores on the fluency test has been raised. From the pre-test (Min= 1.00, Max= 4.00) to the post-test (Min= 3.00, Max= 5.00). Furthermore, there has been a significant increase in the minimum and maximum scores on the pronunciation test. There was an improvement from the pre-test scores (Min= 2.00, Max= 4.00) to the post- test scores (Min= 3.00, Max= 5.00). Finally, the task score range has been raised. Comparing the scores before and after the test showed an improvement. From the pre-test (Min= 2.00, Max= 4.00) to the post-test (Min= 3.00, Max= 5.00). The study uncovered significant enhancements following the implementation of AI as a learning tool for practicing English speaking skills. Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 430 Overall, the findings of this study revealed that AI technology has significantly enhanced the english speaking skills among the EFL students. These results highlighted the potential of AI as a learning tool in speaking skills positively impact language learning outcomes, motivation, and foster the development of abilities and skills among EFL students. Discussion Thematic analysis was employed to investigate the EFL student’s cognition in applying speaking skills through the use of AI (Artificial Intelligence) by semi- structured interview. 3 participants were selected in view of willingness to take part in the interviews regarding this study. Interviewees were coded P1, P2, P3, and so on. The qualitative data from the interviews provided more information and support for the findings above. AI (Artificial Intelligence) has been familiar for the EFL students before. Ling (2023) revealed that participants expressed great enthusiasm for the impact of AI- mediated language training, frequently describing it as a stimulating and engaging learning experience. The users described how the AI platform's interactive exercises, dynamic language games, and instant feedback created an immersive ambiance that kept their attention and made learning pleasurable. This study is aligned with (Terzioglu et al., 2022) stated that, several technologies have been employed by the researcher to overcome the difficulties in speaking where the number of students are encouraged to express themselves openly and embrace the opportunity to learn from their mistakes. The study's findings revealed that students experienced heightened motivation and enhanced self-confidence when interacting in English conversations due to having more opportunities for online practice. In addition, the utilization of speaking skills by English as a Foreign Language (EFL) learners has proven to be a very efficacious learning method. This approach facilitates successful communication among individuals and may be perpetually honed and refined throughout one's lifetime. Crucially, the acquisition of speaking skills is essential for English as a Foreign Language (EFL) learners due to its potential to enhance their oral communication abilities and foster confidence. Moreover, this technology advancement has brought about the advantages in almost all aspects of lives for any almost all ages or young to old generation (Iftanti et al., 2023). EFL students believed AI technology provides assistance and has a positive impact on learning. IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 431 As mentioned by P1: “I believe AI technology can make a positive contribution to language learning by providing personalized feedback.” P2 also added that: “In my opinion, this technology can help us in learning, especially in English, right? For those who are less likely to speak English, AI can help with the work.” P3 additionally noted that: “I think this technology really helps me in learning because it is easier, more practical, and can be studied anywhere. For example, if you have free time at home, you can use it for studying.” The findings from the semi-structured interviews revealed that AI technology has the role to help EFL Students in learning english language, specifically in speaking skills. EFL students found AI to be an effective and efficient in enhancing their English-speaking skills. AI technology improved participants' cognitive abilities and English-speaking skills. One of (Rahmawati et al., 2021) study showed that the way of speaking can be seen through the cognitive strategy. Researchers found data cognitive strategies have a good impact on students. First, practice helps them become more confident in asking for words or commands. Second, the receiving and sending message strategy also positively impacts students, especially in speaking. Third, students learning to speak also carry out analysis and reasoning strategies. Lastly, the strategy of creating a structure for input and output. The results align with prior study on the effects of AI on improving speaking skills (Qiao & Zhao, 2023). According to the semi-structured interviews, the majority of participants concurred that AI had enhanced their cognition in speaking skills including fluency, grammar, pronunciation, vocabulary, comprehension, and task performance. The three participants experienced significant changes before and after using AI. Keong et al., (2015) investigated speaking competence of Iraqi EFL undergraduates of Garmiyan University. The findings of the study indicated that participants faced some challenges in relation to their linguistics competences, such as vocabulary, Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 432 grammar, and pronunciation. The study also revealed that the lack of English- speaking practice, the tendency of mixing the language, and the absence of speaking activity in the course program are the main reasons leading to EFL students’ speaking difficulties. Before using AI, the EFL students realized that they found difficulties in applying english speaking skills. As indicated by P1: “Before using AI, you could only learn by reading, playing games, and listening to music in English.” As stated by P2: “Before using AI, it was possible to understand about english, but it still needs more practice.” The participant stated that before using AI, she was able to understand english. However, she couldn’t really understand. P3 added that: “Before using AI, I only use simple methods such as speaking in front of the teacher.” This indicated that participants could only accessed minimal methods to practice speaking skills before using AI technology. Normawati et al., (2023) demonstrated that there are three main difficulties faced by the students in developing their English-speaking skills. The first difficulty deals with the aspect of grammar, which is caused by the differences between the students’ first language and English, the language being learnt. The second issue is difficulty with pronunciation in English due to the inhibition of the English pronunciation system. Therefore, it is crucial for EFL students to apply another alternative method to develop their cognition in english speaking skills. In contrast, most participants asserted that significant improvements have been observed in utilizing AI for English speaking. Previous study has shown that, AI severs as a valuable and enjoyable educational tool for enhancing english language proficiency (Agustini, 2023). EFL students have acquired several cognitive skills through the use of AI such as grammar, vocabulary, fluency, comprehension, pronunciation, and tasks. However, The EFL students were most adept at understanding pronunciation and fluency in terms of speaking skills. For instance, P1 stated that: “After using AI, my knowledge has expanded through significant understanding. IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 433 For example, I came to know the extent of my abilities in pronunciation, fluency, and so on. We are also given the correct answer if the AI detects the wrong answer.” P2 also added that: “Because after using AI, I can understand english language better. Because, AI can help too. Like, it could improve my pronunciation. Such as pronunciation, intonation or grammar.” Furthermore, P3 pointed out that: “After using AI, I know more about the extent of my speaking skills in using English, including the accurate pronunciation” The participants specifically thought that pronunciation and fluency were the skills that made the most progress in. The EFL students were most adept at understanding pronunciation and fluency in terms of speaking skills. According to a recent study, most participants rely on technology to enhance their English skills, including pronunciation (Adara et al., 2023). Mulyani & Sartika (2019) emphasized that pronunciation is crucial for attaining communication competence. One study suggests that AI can offer immediate feedback on pronunciation accuracy, assisting learners in improving their spoken language skills (Saraswati et al., 2023). The current research aligns with the findings of Karim et al. (2023), highlighting that most participants believe that AI can assist students in pronouncing words accurately, boosting their confidence during speaking sessions, and recognizing their mistakes. Dealing with the fluency skills, a study by Iswara et al. (2012) found that creating short turns in communication improved students' fluency skills. This led to better questioning, responding, answering questions, and exchanging ideas. Being fluent means being able to speak without any interruptions (Murti et al., 2022). To investigate a further answer regarding the utilization of AI in english speaking skills, the participants added their opinion about the cognition of AI improvements’ in english speaking skills. The understanding of the enhancements in speaking skills that EFL students have acquired greatly advanced. As stated by P1: “So, the use of AI can provide improvement to my speaking skills to target practice and feedback that help improve my skills. OK, so I have had many Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 434 weaknesses until now. It's just that over time I've been using AI, it's like, for example, those verbs. If AI helps, it's starting to get easier now. There are judgments too. For example, the AI tells you the tone should be like this, then uses this, saying the intonation is not this one, but this one. There is improvement in pronunciation, fluency, etc.” P2 additionally noted that: ” There is improvement. Maybe you can improve your pronunciation, you can repeat it, but not yet intonation. My improvement in speaking is more pronunciation skills.” Furthermore, P3 noted that: “I am capable to pronounce the sentence correctly. What was previously incorrect means I could say it better. There is also improvement in grammar and vocabulary. From the interviews above, this means that AI has the ability to help the improvements of speaking skills to the EFL students. AI can provide assistance to facilitate EFL students' learning in speaking skills. It was shown that technology including AI is capable of improving English language proficiency. Many individuals acknowledged significant enhancement in their ideas after practicing spoken English with the AI. Rusmiyanto et al., (2023) assert that AI (Artificial Intelligence) could give the positive impact for EFL students on improving communication skills and pronunciation skills. In addition, empirical evidence from pre-test and post-test studies has demonstrated significant improvements in the speaking skills of EFL students, involving grammar, vocabulary, fluency, comprehension, pronunciation, and tasks. The paired sample t-test data indicates a significant discovery that is both theoretically and practically important (Al-Khresheh, 2024). This discovery aligns with prior studies investigating the extent to which EFL students can comprehend and enhance different speaking skills by utilizing AI (Madhavi et al., 2023). Zou et al., (2023) assert that AI helped students to practice speaking performance such as grammar, vocabulary, fluency, and pronunciation. Another intriguing finding indicated that the majority of participants had positive responses towards AI as a learning tool in speaking skills independently (Li & Zou, 2022). Similarly, Obari et al., (2022) research explored that AI improved the EFL students' overall English ability. It was also discovered that the post-training survey findings showed that participants found employing an AI to be an enjoyable, user-friendly, and effective method for enhancing their English speaking and vocabulary skills. According to IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 435 (Ericsson and Johansson, 2023), in their research about “English speaking practice with conversational AI: Lower secondary students’ educational experiences over time”, a conversational AI was utilized where students conveyed that the exercises provided in the system were effectively interconnected and relevant to the topic. The students began to gain expertise and found it increasingly effortless to operate within the system. In order to serve the growing demand for education, it is imperative that we create and use artificial intelligence solutions to support the learning requirements of our students. Integrating AI into education is essential for implementing a well-planned method to deliver high-quality education (Kumar. 2023). Conclusion This study explored the utilization of Artificial Intelligence (AI) to improve English speaking skills and enhance the cognitive abilities of EFL students. The findings indicated that AI serves as an educational tool that can assist students in strengthening various speaking abilities. There was a notable improvement in each skill related to English speaking. The most improved skills were grammar, fluency, and pronunciation. EFL students identified pronunciation and fluency as the most commonly highlighted qualities in speaking throughout the semi-structured interview. Moreover, AI was seen as a proficient and efficient tool aiding English communication. The findings showed that AI is viewed as an enjoyable, convenient, and effective option for improving English speaking skills. There are limitations in the current study. The investigation lasted only 2 weeks and was conducted during teaching and learning activities. Furthermore, students had limited opportunity to engage in this experiment as it only lasted 2 hours of learning. Most students' gadgets cannot contact the AI due to limited memory capacity. Hence, the sample size needed to be increased to represent the results adequately. Future research endeavors may benefit from an extended duration and the utilization of a substantial sample size for the experiment. This work may offer valuable knowledge and perspectives for EFL students seeking to utilize AI for educational purposes. The results of this study can offer useful insight into the efficacy of utilizing AI to enhance English speaking abilities in EFL students. This research aims to provide recommendations and guidance for developing more effective AI technology in English language learning in the future by studying how students use AI technology. All students from various fields are encouraged to Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 436 use AI as a supporting tool in their learning activities. References Adara, R. A., Zulaeha, E. R., Arianto, A., Kamilah, W., & Najmudin, O. (2023). Analyzing the correlation between students’ motivation and error pronunciation of voiceless Θ words: a Qualitative study. Eralingua, 7(2), 377. https://doi.org/10.26858/eralingua.v7i2.46793 Alfatihah, A., Ismayanti, D., Syam, A., & Santaria, R. (2022). Teaching Speaking Skills through Project-Based Learning for the Eighth Graders of Junior High School. IDEAS: Journal on English Language Teaching and Learning, Linguistics and Literature, 10(1), 152-165. doi:https://doi.org/10.24256/ideas.v10i1.2555 Agustini, N. P. O. (2023). Examining the Role of ChatGPT as a Learning tool in Promoting Students' English Language Learning Autonomy relevant to Kurikulum Merdeka Belajar. EDUKASIA: Jurnal Pendidikan dan Pembelajaran, 4 (2), 921-934. http://jurnaledukasia.org Al-Khresheh, M. H. (2024). The role of presentation-based activities in enhancing speaking proficiency among Saudi EFL students: A quasi-experimental study. Acta Psychologica, 1-10. https://doi.org/10.1016/j.actpsy.2024.104159 Brown. H. D. 2001. Teaching by Principles: An Interactive Approach in Language Pedagogy. (2nd Ed). New York: Pearson Education co. Braun, V., & Clarke, V. (2006). Using Thematic Analysis in Psyhology. Qualitative Research in Psychology, 3 (2), 77-101. https://doi.org/10.1191/1478088706qp063oa Campbell, D. T., & Stanley, J. (1963). Experimental and quasi-experimental designs for research. Chicago, IL: Rand McNally. Creswell, J. W. (2012). Educational research: Planning, conducting, and evaluating quantitative and qualitative research (4th ed.). Boston, MA: Pearson. Creswell, J. W. (2014). Research Design: Qualitative, Quantitative and Mixed Methods Approaches (4th ed.). Thousand Oaks, CA: Sage. Ericsson, E., &Johansson, S. (2023). English speaking practice with conversational AI: Lower secondary students’ educational experiences over time. Computers and Education: Artificial Intelligence, 1-13. https://doi.org/10.1016/j.caeai.2023.100164 https://doi.org/10.26858/eralingua.v7i2.46793 http://jurnaledukasia.org/ https://doi.org/10.1016/j.actpsy.2024.104159 https://doi.org/10.1191/1478088706qp063oa https://doi.org/10.1016/j.caeai.2023.100164 IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 437 Gayed, J. M., Carlon, M. K. J., Oriola, A. M., & Cross, J. S. (2022). Exploring an AI-based writing Assistant’s impact on English language learners. Computers and Education: Artificial Intelligence, 1-7. https://doi.org/10.1016/j.caeai.2022.100055 Hulsbergen, M. H., Jong, J. D., & Smagt, M. J. V. D. (2022). Exploring the use of Online Simulations in Teaching Dialogue Skills. Psychology Learning and Teaching, 1(22), 55-73. https://doi.org/10.1177/14757257221138936 Iftanti, E., Awalin, A. S., & Izza, F. N. (2023). THE USE OF ARTIFICIAL INTELLIGENCE AS THE POTENTIAL SUPPORTING LEARNING TOOLS FOR DOING LEARNING PROJECTS. The Ananual International Conference on education, 1, 455-467. Iswara, A. A., Azib, A., & Rochsantiningsih, D. (2012). IMPROVING STUDENTS’ SPEAKING FLUENCY THROUGH THE IMPLEMENTATION OF TRIVIA- BASED ACTIVITY IN UNIVERSITY STUDENTS. English Education: Jurnal Pendidikan Bahasa Inggris Universitas Sebelas Maret, 1 (1), 1-17. Karim, S. A., Hamzah, A. Q. S., Anjani, N. M., Prianti, J., & Sihole, I. G. (2023). Promoting EFL Students’ Speaking Performance through ELSA Speak: An Artificial Intelligence in English Language Learning. JOLLT Journal of Language and Language Teaching, 11 (4), 655-668. https://doi.org/10.33394/jollt.v%vi%i.8958 Kaur, P., Jill, S., & Vikas, Y. (2018). Descriptive Statistics. International Journal of Academic Medicine, 4(1), 60-63. DOI: 10.4103/IJAM.IJAM_7_18 Keong, Y. C., Ali, A. I., & Hameed, F. W. (2015). Speaking competence of Iraqi EFL undergraduates of Garmiyan University. International Journal of Education and Research, 3(5), 157-170 Khosravi, H., Shum, S. B., Chen, G., Conati, C., Tsai, Y. S., Kay, J., Knight, S., Maldonado, R. M., Sadiq, S., & Gasevic, D. (2022). Explainable Artificial Intelligence in Education. Computers and Education: Artificial Intelligence, 1-22. https://doi.org/10.1016/j.caeai.2022.100074 Kim, K., & Kwon, K. (2023). Exploring the AI competencies of elementary school teachers in South Korea. Computers and Education: Artificial Intelligence, 1-11. https://doi.org/10.1016/j.caeai.2023.100137 Kumar, M. J. (2023). Artificial Intelligence in Education: Are we ready? ETE TECHNICAL REVIEW, 40 (2), 153-154. https://doi.org/10.1080/02564602.2023.2207916 https://doi.org/10.1016/j.caeai.2022.100055 https://doi.org/10.1177/14757257221138936 https://doi.org/10.33394/jollt.v%25vi%25i.8958 https://doi.org/10.1016/j.caeai.2022.100074 https://doi.org/10.1016/j.caeai.2023.100137 https://doi.org/10.1080/02564602.2023.2207916 Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 438 Li, M., Zou, B. (2022). Using artificial intelligence technology as a tutor for EFL speaking practice. In Smart CALL: Personalization, Contextualization, Socialization; Colpaert, J., Stockwell, G., Eds.; Castledown: London, UK, 82– 100.DOI: 10.29140/9781914291012-5 Ling, W. (2023). Artificial intelligence in language instruction: impact on English learning achievement, L2 motivation, and self-regulated learning. Frontiers in Psychology, 14. https://doi.org/10.3389/fpsyg.2023.1261955 Madhavi, E., Sivapurapa, L., Koppula, V., Rani, P. B. E., & Sreehari, V. (2023). Developing Learners’ English Speaking Skills using ICT and AI Tools. Journal of Advanced Research in Applied Sciences and Engineering Technology, 142-153. https://doi.org/10.37934/araset.32.2.142153 Maguire, M., & Delahaunt, B. (2017). Doing a Thematic Analysis: A Practical, Step- by-Step Guide for Learning and Teaching Scholars. All Ireland Journal of teaching and Learning in Higher Education (AISHE-J), 8 (3), 3351-33514. http://ojs.aishe.org/index.php/aishe-j/article/view/335 Masuram, J., & Sripada, P. N. (2020). DEVELOPING SPEAKING SKILLS THROUGH TASK-BASED MATERIALS. Procedia Computer Science, 60-65. https://doi.org/10.1016/j.procs.2020.05.009 Masruddin, M., & Nasriandi, N. (2022). Lexical and Syntactical Errors Performed by Junior High School Student in Writing Descriptive Text. IDEAS: Journal on English Language Teaching and Learning, Linguistics and Literature, 10(1), 1094-1100. doi:https://doi.org/10.24256/ideas.v10i1.3024 Masruddin, M., Amir, F., Langaji, A., & Rusdiansyah, R. (2023). Conceptualizing linguistic politeness in light of age. International Journal of Society, Culture & Language, 11(3), 41-55. Mihas, P. (2023). Thematic Analysis. International Encyclopedia of Education (Fourth Edition), 2023 Mulyani, M., & Sartika, D. (2019). ANALYSIS OF STUDENTS’ PROFICIENCY ON ENGLISH PRONUNCIATION USING YOUTUBE-BASED VIDEO MEDIA. English Education Journal, 6(2), 248– 257. https://doi.org/10.46244/geej.v6i2.881 Murti, P., Jabu, B., & Samtidar. (2022). STUDENTS’ ENGLISH SPEAKING DIFFICULTIES AND TEACHERS’ STRATEGIES IN ENGLISH TEACHING: A CASE STUDY AT SMA NEGERI 10 GOWA. PERFORMANCE: Journal of English Education and Literature, 1 (1), 50-63. https://ojs.unm.ac.id/performance https://doi.org/10.3389/fpsyg.2023.1261955 https://doi.org/10.37934/araset.32.2.142153 http://ojs.aishe.org/index.php/aishe-j/article/view/335 https://doi.org/10.1016/j.procs.2020.05.009 https://www.sciencedirect.com/science/article/pii/B978012818630511036X https://www.sciencedirect.com/science/article/pii/B978012818630511036X https://doi.org/10.46244/geej.v6i2.881 https://ojs.unm.ac.id/performance IDEAS, Vol. 12, No. 1, June 2024 ISSN 2338-4778 (Print) ISSN 2548-4192 (Online) 439 Normawati, A. Nugrahaeni, D. A. Manggolo, N. S. K. H. & Susanto, A. I. F. (2023). EFL Learners’ Difficulties in Speaking English. The English Language and Education System, 1 (1), 1-9. https://doi.org/10.53416/electrum.v1i1.116 Obari, H., Lambacher, S. G., & Kikuchi, H. (2022). Exploring the impact of AI on EFL teaching in Japan. In Castledown Publishers eBooks, 101–125. https://doi.org/10.29140/9781914291012-6 Qiao, H., & Zhao, A. (2023). Artificial intelligence-based language learning: illuminating the impact on speaking skills and self-regulation in Chinese EFL Context. Frontiers in Psychology, 1-15. https://doi.org/10.3389/fpsyg.2023.1255594 Rahmawati, S., Akib, E., & Ilmiah. (2021). THE STUDY OF COGNITIVE STRATEGIES IN LEARNING SPEAKING USED BY THE THIRD SEMESTER STUDENTS OF ENGLISH EDUCATION DEPARTMENT AT ALAUDDIN STATE ISLAMIC UNIVERSITY. Journal of Language Testing and Assessment, 1 (2), 164-169. https://jurnal.fkip.unismuh.ac.id/index.php/jlta/index Rusmiyanto, R., Huriati, N., Fitriani, N., Tyas, N., Rofi’i, A., & Sari, M. (2023). The Role Of Artificial Intelligence (AI) In Developing English Language Learner’s Communication Skills. Journal on Education, 6(1), 750- 757. https://doi.org/10.31004/joe.v6i1.2990 Saraswati, G. P. D., Farida, A. N., & Yuliati. (2023). Implementing AI Replika In Higher Education Speaking Classes: Benefits and Challenges. Journal of English Language Teaching, 12 (3), 207-215. http://journal.unnes.ac.id/sju/index.php/elt Southworth, J., Migliaccio, K., Glover, J., Glover, J., Reed, D., McCarty, C., Brendemuhl, J., & Thomas, A. (2023). Developing a Model for AI Across the Curriculum: Transforming the higher education landscape via innovation in AI literacy. Computers and Education: Artificial Intelligence, 1-10. https://doi.org/10.1016/j.caeai.2023.100127 Tajik, O., &Golzar, J. (2022). Simple Random Sampling. International Journal of Education and Language Studies, 1 (2), 78-82. DOI:10.22034/ijels.2022.162982 Terzioglu, Y., & Kurt, M. (2022). Elevating English Language Leaners’ Speaking Fluency and Listening Skill Through A Learning Management System. Original Research, 12 (2), 1-15. DOI: 10.1177/21582440221099937 https://doi.org/10.53416/electrum.v1i1.116 https://doi.org/10.29140/9781914291012-6 https://doi.org/10.3389/fpsyg.2023.1255594 https://jurnal.fkip.unismuh.ac.id/index.php/jlta/index https://doi.org/10.31004/joe.v6i1.2990 http://journal.unnes.ac.id/sju/index.php/elt https://doi.org/10.1016/j.caeai.2023.100127 http://dx.doi.org/10.22034/ijels.2022.162982 Namira Fasya Khailifa , Pirman Ginting Exploring EFL Students’ Cognition and Practices in Applying English Speaking Through the Use of AI (Artificial Intelligence) 440 Zou, B., Du, Y., Wang, Z., Chen, J., & Zhang, W. (2023). An Investigation Into Artificial Intelligence Speech Evaluation Programs With Automatic Feedback For Developing EFL Learners’ Speaking Skills. Sage Open, 1-9. https://doi.org/10.1177/21582440231193818 Zou B, Guan X, Shao Y, Chen P. Supporting Speaking Practice by Social Network- Based Interaction in Artificial Intelligence (AI)-Assisted Language Learning. Sustainability. (2023), 15(4), 28-72. https://doi.org/10.3390/su15042872 https://doi.org/10.3390/su15042872