Maestro y Sociedad e-ISSN 1815-4867

Volume 23 Number 1 Year 2026

Original article

Strategy to improve fluency in English

Estrategia para mejorar la fluidez en el idioma inglés

Estratégia para melhorar a fluência na língua inglesa

Estudiante Bianca Hidali Pisco Tuarez *, https://orcid.org/0009-0009-2984-047X

Estudiante Diana Carolina Navarrete Parraga, https://orcid.org/0009-0002-8321-3855

Raisa Macias Sera, https://orcid.org/0000-0001-9618-2854

Universidad Laica “Eloy Alfaro” de Manabí Extensión Chone, Ecuador

*Corresponding author. email e1752168979@live.uleam.edu.ec

To cite this article: Pisco Tuarez, B. H., Navarrete Parraga, D. C. & Macias Sera, R. (2026). Strategy to improve fluency in English. Maestro y Sociedad, 23(1), 340-349. https://maestroysociedad.uo.edu.cu

Abstract

Introduction: Oral fluency in English is a multifactorial challenge that integrates linguistic and affective aspects. In Ecuador, low levels of English proficiency demonstrate deficiencies in this area. The objective of this study was to design a teaching strategy supported by artificial intelligence (AI) to improve fluency in high school students. Materials and methods: This mixed-methods, correlational study employed inductive-deductive methods. The sample included 30 students from the Cinco de Mayo Educational Unit. Observation sheets and teacher interviews were used for the diagnostic assessment. Results: The results revealed significant limitations in fluency, restricted vocabulary, and insecurity when speaking in public. In response, a four-stage strategy was designed centered on the Sesame AI platform, which offers a private practice environment, immediate corrective feedback, and reduces the affective filter. Discussion: The strategy was positively evaluated by teachers during its implementation, who considered it motivating and effective in providing emotional and technical support, aligning with the literature that highlights the potential of AI to personalize learning and foster autonomy. Conclusions: It is concluded that the integration of AI, through a structured pedagogical strategy, constitutes a viable tool for enhancing fluency by combining recurrent language practice with affective support that mitigates anxiety, without replacing the teacher's role.

Keywords: strategy, artificial intelligence, fluency, English language learning.

Resumen

Introducción: La fluidez oral en inglés es un desafío multifactorial que integra aspectos lingüísticos y afectivos. En Ecuador, los bajos niveles de dominio del inglés evidencian deficiencias en esta área. El objetivo de este estudio fue diseñar una estrategia didáctica apoyada en inteligencia artificial (IA) para mejorar la fluidez en estudiantes de bachillerato. Materiales y métodos: La investigación, de enfoque mixto y nivel correlacional, empleó métodos inductivo-deductivo. La muestra incluyó 30 estudiantes de la Unidad Educativa Cinco de Mayo. Se utilizaron fichas de observación y entrevistas a docentes para el diagnóstico. Resultados: Los resultados revelaron limitaciones significativas en la fluidez, vocabulario restringido, e inseguridad al expresarse en público. En respuesta, se diseñó una estrategia de cuatro etapas centrada en la plataforma Sesame AI, la cual ofrece un entorno de práctica privado, retroalimentación correctiva inmediata y reduce el filtro afectivo. Discusión: La estrategia fue valorada positivamente por los docentes durante su socialización, quienes la consideraron motivadora y eficaz para proporcionar apoyo emocional y técnico, coincidiendo con la literatura que destaca el potencial de la IA para personalizar el aprendizaje y fomentar la autonomía. Conclusiones: Se concluye que la integración de la IA, a través de una estrategia pedagógica estructurada, constituye una herramienta viable para potenciar la fluidez, al combinar la práctica lingüística recurrente con un soporte afectivo que mitiga la ansiedad, sin sustituir la labor docente.

Palabras clave: estrategia, inteligencia artificial, fluidez, aprendizaje de inglés.

Resumo

Introdução: A fluência oral em inglês é um desafio multifatorial que integra aspectos linguísticos e afetivos. No Equador, os baixos níveis de proficiência em inglês demonstram deficiências nessa área. O objetivo deste estudo foi desenvolver uma estratégia de ensino apoiada por inteligência artificial (IA) para melhorar a fluência de alunos do ensino médio. Materiais e métodos: Este estudo correlacional de métodos mistos empregou métodos indutivo-dedutivos. A amostra incluiu 30 alunos da Unidade Educacional Cinco de Mayo. Fichas de observação e entrevistas com professores foram utilizadas para a avaliação diagnóstica. Resultados: Os resultados revelaram limitações significativas na fluência, vocabulário restrito e insegurança ao falar em público. Em resposta, foi desenvolvida uma estratégia em quatro etapas centrada na plataforma Sesame AI, que oferece um ambiente de prática privada, feedback corretivo imediato e reduz o filtro afetivo. Discussão: A estratégia foi avaliada positivamente pelos professores durante sua implementação, que a consideraram motivadora e eficaz no fornecimento de suporte emocional e técnico, alinhando-se à literatura que destaca o potencial da IA para personalizar o aprendizado e promover a autonomia. Conclusões: Conclui-se que a integração da IA, por meio de uma estratégia pedagógica estruturada, constitui uma ferramenta viável para aprimorar a fluência, combinando a prática recorrente da língua com o apoio afetivo que atenua a ansiedade, sem substituir o papel do professor.

Palavras-chave: estratégia, inteligência artificial, fluência, aprendizagem de inglês

Received: 7/11/2025 Approved: 6/1/2026

Introduction

Traditionally mediated by structural or traditional methods and subject to face-to-face learning, language learning in educational contexts has undergone a significant transformation due to the incorporation of emerging technologies such as Artificial Intelligence (AI), which has gained prominence for its ability to offer adaptability, immediate corrective intervention, and personalization of the teaching-learning process. This technological resource has proven useful in promoting autonomous, flexible, and interactive learning, and has positioned itself as a relevant tool in the contemporary educational field. (López Regalado et al., 2024)

In the process of teaching and learning foreign languages, the development of oral skills is considered one of the greatest challenges for both teachers and students. Chamorro Ortega et al. (2020) argue that fluency, as an important aspect of oral skills, involves multiple components: from the number and length of pauses to accuracy, promptness, reformulation, and repetition of words. These elements, together with factors such as self-perception of one's own ability and confidence when speaking, constitute a complex dimension of language learning. Skehan (2009) reinforces this perspective by pointing out that fluency requires not only grammatical and lexical knowledge, but also the ability to use that knowledge naturally in real communicative situations.

Research related to the approach to English language teaching in Ecuador shows concern about the results obtained to date. According to Education First (2025) the country ranks 82nd, placing it among those with the lowest linguistic competence in this language in Latin America. Despite the government's efforts through the Ministry of Education to improve the quality of English language teaching through educational programs such as “Ecuador habla inglés” (Ecuador speaks English) and SIT TESOL Best Practices, endorsed by the U.S. Embassy, there is still a lack of continuity, follow-up, and adaptation to real contexts, which has limited the effective development of oral skills, and although there has been progress, traditional methodologies and materials isolated from context hinder meaningful language acquisition.

An example of the shortcomings in English language learning in the country, and specifically in Chone, was observed by the authors of this research during their pre-professional internships at the Cinco de Mayo Educational Unit, where they found deficiencies in oral skills, specifically in fluency during oral activities in English. Among the shortcomings to be highlighted are a lack of naturalness when speaking, preventing spontaneous use of the language, a limited vocabulary, and prolonged pauses when speaking, among others. Given the problem presented, the limitations in oral fluency in English that hinder the learning of high school students were declared as a research problem, and the general objective was to design a strategy to enhance fluency in English learning for high school students. The proposed strategy is characterized by three closely related stages and the use of AI-supported technology. The SESAME AI platform was selected for the development of the research, and a sample of 30 students was taken from the Cinco de Mayo Educational Unit in the canton of Chone, as this was the class group with which the pre-professional practice was carried out.

The research is useful because it allows students to develop a safe environment that reduces anxiety and fear of making mistakes, enabling them to practice pronunciation and oral expression in a private and personalized manner. On the other hand, for teachers, AI acts as a fundamental teaching assistant that allows them to supervise the individual practice of large groups of students simultaneously, providing accurate data on the progress and specific difficulties of each student.

Artificial Intelligence (AI)

Gupta (2025) defines artificial intelligence as a simulation of human intelligence, capable of simulating cognitive processes and adapting to different contexts through algorithms and advanced computer systems. Its evolution has followed an upward and exponential trajectory; its application has extended beyond the social sphere to become significantly integrated into the educational and professional sectors, transforming the way we learn and work.

Along the same lines, data-based AI has become more relevant, relying on the analysis of large volumes of information, which are processed using machine learning algorithms. Through this process, the system is able to identify patterns, trends, and relationships within the data, allowing it to generate responses and make decisions autonomously. This is considered one of the most popular types of AI, largely due to the exponential growth of available data and advances in computational capabilities that enable its efficient processing. (López de Mántaras & Brunet Crosa, 2023)

AI in education

According to the study by Mendoza & Torres Zapata (2024) the use of Information and Communication Technology (ICT) in education is significantly valued by teachers, with 66.7% of women and 70.0% of men agreeing. This group has seen how these tools have fostered motivation among students, highlighting that access to a greater number of resources generates a pedagogical advantage. As a result, the teaching-learning process is being redefined, offering new platforms with tools for the development of educational skills. Asensio Iglesias (2024) demonstrates that tools such as ChatGPT, Duolingo, Character AI, and ELSA Speak allow for the personalization of content and the adaptation of learning to the individual styles and rhythms of students. These platforms have been effective not only in improving student motivation and autonomy, but also in optimizing academic performance through task automation and continuous feedback.

AI in English language learning

The implementation of this technological tool has begun to streamline the teaching-learning process, integrating digital instruments that enhance the acquisition of language skills. According to Ayala & Alvarado (2023) the growing demand for competent teachers with truly developed linguistic skills and sub-skills drove the creation of tools to improve and accelerate the English language learning process. Under this premise, AI manifests itself through various platforms using algorithms to identify students' strengths and weaknesses in the language.

AI in English language learning offers multiple benefits, allowing for unprecedented personalization, adapting to each student's learning style, facilitating access to high-quality educational resources anytime, anywhere, democratizing learning, and even reducing the workload of teachers Chen et al. (2020) proof of this is the use of AI systems such as ChatGPT, specifically the CHOP platform, which generates personalized feedback in real time for oral presentations, helping to improve the accuracy, intonation, and coherence of speech in EFL students. (Cha et al., 2024)

AI platforms that enhance fluency in English learning

According to Martínez et al. (2023) the integration of technology in language education has led to the development of various virtual tools that help students strengthen their language skills, offering interactive and personalized environments that adapt to the individual needs of each user, thus facilitating effective and accessible learning.

Among the most notable platforms are the following:

Yoodli: Considered an AI speech coach designed to improve fluency in public speaking and everyday conversations. Users can practice speeches, presentations, or informal conversations, and it offers specific suggestions for improvement by AI algorithms to analyze speech, providing instant and personalized feedback on aspects such as speech rate, use of filler words, clarity, intonation, and pronunciation. (Deepgram, 2025)

Character AI: An AI platform that acts as characters with defined personalities, whether created by other users or by yourself, and is designed for more creative, emotional, or role-play interactions, allowing you to explore stories, simulate situations, or simply entertain yourself.

Following the diagnosis, which identified a marked lack of fluency and a reliance on monosyllabic responses, the researchers selected the Sesame AI platform, which features characters such as Maya and Miles who facilitate a much more natural and empathetic interaction. This technological evolution, detailed by experts at Salesians Sarria, is based on the Conversational Speech Model, allowing AI to be much more capable of answering questions, participating in fluid dialogues by adjusting tone, pauses, and rhythm to the student's context. By using a realistic human voice, with pauses, emotions, and language close to that of a native speaker, it reinforces knowledge through constant dialogue and immediate feedback, simulating the experience of studying with a trusted friend. (Jordi, 2025)

The central purpose of Sesame AI is to improve the quality of oral communication, making conversations feel human and less artificial, thus being perfect for shy learners who avoid direct interaction. Its main function in this would be to facilitate language practice, resolve doubts, adapt to the user's pace, and provide instant feedback in a natural environment. (AI, s. f.)

This system transcends traditional tutoring by offering a safe space, reducing academic anxiety, and promoting mastery of content similar to that of a human tutor through constant, two-way interaction.

Fluency as an aspect of oral skill in foreign language teaching

Fluency in language learning has been characterized by the ability to speak extensively with a minimum of empty pauses in inappropriate positions. There is also talk of incorporating characteristics of continuous speech, without denying the relevance of speech speed and rhythm, where the duration of execution between pauses are key indicators of the same skill. Likewise, prosody includes the correct use of intonation and accent for the benefit of easy interpretation and communication in conversation. All these characteristics come together to give rise to fluency, understood as a skill within communicative competence, which allows the speaker to maintain coherent and appropriate discourse production in real contexts. (Uribe et al., 2024)

Segalowitz (2005) introduces the concept of fluency, understood as the ability to produce utterances quickly and effortlessly, the result of constant and structured practice. This linguistic automation frees up cognitive resources for content and interaction, and according to Grasso (2024) it demonstrates an automatic process of language planning and execution, as he defines it as the ability to simultaneously conceive and express oneself. Strengthening fluency in the foreign language classroom involves simultaneously developing these cognitive, lexical, and discursive dimensions through meaningful tasks, sustained interaction, and continuous feedback, in order to promote spontaneous and comprehensible production.

Aspects of fluency

Segalowitz (2010) presents a three-part taxonomy of fluency: Cognitive fluency allows ideas to be organized and statements to be formulated quickly and coherently, which is facilitated by familiarity with syntactic structures and a broad lexical repertoire. Statement fluency is related to observable variables such as articulation speed, the number and type of pauses, repetitions, and hesitations; perceived fluency is linked to the listener's impression of the clarity, naturalness, and rhythm of the speaker's discourse.

Aspects of fluency in English as a foreign language

Formulaic sequences and perceived fluency

Formulaic sequences, such as frequently used pre-constructed expressions (“you know,” “at the end of the day,” “in other words”), play a central role in the development of fluency, especially in English as a foreign language.

These structures reduce cognitive load, as they do not require active grammatical or lexical planning at the moment of speaking, which improves both the fluency of the utterance and the fluency perceived by the listener. According to Boers et al. (2006) students who incorporate formulaic sequences into their speech are evaluated as more fluent and natural, even when their overall vocabulary is limited. Likewise, Lambert et al. (2021) state that the presence of these sequences predicts a greater perception of fluency by external evaluators.

Pauses and clause breaks

One of the most obvious features of lack of fluency in English learners is the excessive use of pauses within clauses, which interrupts the flow of speech. Pauses in syntactically inappropriate positions (e.g., in the middle of a spoken sentence) are interpreted as signs of hesitation or cognitive difficulties in processing language. In this regard, Uribe et al. (2024) point out that these interruptions, called “unconnected pauses,” are a direct indicator of a lack of automation in speech processes, as they occur outside the expected syntactic boundaries. In addition, it has been observed that native English speakers tend to pause at natural syntactic boundaries, while learners pause more frequently within clauses, which affects the naturalness of speech.

Prosody and intonation

Prosody, understood as the set of patterns of rhythm, stress, and intonation, is also an essential component of fluency. In the case of English, the proper use of stress and melodic patterns not only contributes to the clarity of the message, but also to the perception of oral competence. Jong & Wempe (2009) highlight that controlled prosody correlates with better scores in fluency assessments.

Cognitive fluency and processing speed

Cognitive fluency, which refers to the speed with which the speaker retrieves words and grammatical constructions during oral production, is key to achieving spontaneous fluency. In students who are learning English as a foreign language, this type of fluency develops with the automation of collocations, grammatical structures, and syntactic patterns. Romero et al. (2019) confirm that fluency is a manifestation of executive functioning that requires not only access to the lexicon, but also cognitive organization skills, focused attention, and the ability to search for words strategically. This process depends on the efficiency of brain networks and functions such as working memory, which allow information stored in long-term memory to be retrieved quickly when faced with novel or demanding tasks.

ICT-based strategies for fluency

Mobile-Assisted Language Learning (MALL)

Mobile-Assisted Language Learning (MALL) has established itself as an effective strategy for improving oral skills. According to a study by Salih & Omar (2024) this methodology promotes ubiquitous access to learning, increases student motivation, and fosters autonomy. The impact of MALL on speaking skills: their literature review highlights more oral interactions, personalized modules, and peer collaboration, which contribute to progress in fluency.

Interaction between ICT and fluency development

The use of instant feedback applications, powered by AI and speech recognition, facilitates real-time pronunciation and intonation correction, contributing to automatic fluency. In this regard, Revelo Trujillo & Ramirez Roman (2024) argue that the integration of digital tools and self-regulated cognitive strategies allows students to optimize their oral competence through self-correction and constant repetition. In addition, the flexibility of mobile learning allows for the integration of fluency into the student's daily routine, increasing active exposure time to the language, and strengthening the cognitive processes involved in fluency.

Methodology

To conduct this research, a mixed approach was used which, according to Romero et al. (2023) integrates quantitative and qualitative methods to offer a broader and deeper view of social phenomena, allowing for the analysis of both numerical data and the meanings and experiences that accompany them, which promotes a holistic understanding of reality. This combination facilitates a broader and deeper understanding of the educational realities experienced at the Cinco de Mayo Educational Unit in relation to teaching and improving English language fluency.

The qualitative approach was applied through semi-structured interviews with teachers and an observation sheet during English classes. These instruments provided insight into the perceptions, practices, and experiences related to the pedagogical strategies implemented to develop fluency in students.

The descriptive level, primarily by collecting and organizing information on the strategies applied to promote fluency in English, included the attitudes and opinions of the educational actors involved. Through interviews with teachers, classroom observations, and student surveys, a detailed view of the English teaching-learning process was obtained, focusing especially on oral skills. This level allowed for the objective recording of current practices, laying the foundation for further analysis.

The explanatory or correlational level was also considered in order to analyze possible relationships between relevant variables, such as the frequent use of oral practice strategies (such as role-playing, guided conversations, or repetition) and the level of confidence or performance perceived by students when speaking English. Although the aim is not to establish definitive causal relationships, this level allows for the identification of patterns that contribute to a clearer understanding of the impact of the strategies used. The findings will enable the generation of pedagogical recommendations to help improve English fluency in the classroom.

The population consisted of 221 students from the Cinco Mayo Educational Unit. A sample of 30 students from parallel high school “A” was taken, where instruments such as teacher interviews and classroom observation were applied.

A fundamental contribution of the research was the design of a strategy to promote English language fluency among high school students. This strategy consists of four stages, each of which explains the actions to be carried out by teachers and students with the support of the SESAME AI.

Results

Results of the observation

The analysis of the observation carried out in the second year of high school reveals a significant contrast between the students' attitudes and their actual language skills. Although effective practices have been identified in terms of organization and respect for rules of coexistence, there are critical challenges in oral production that directly impact the quality of English language learning. The most relevant aspects observed are detailed below.

Fluency and Limitations in Expression

There is a marked lack of fluency, as students rarely manage to speak without using filler words. Their responses are often limited to isolated sentences or single words, showing an inability to expand on ideas spontaneously. The pace of speech is very close to that of their mother tongue, resulting in conversations that are unexpressive and mechanical. In addition, there is a significant challenge in the emotional realm, as students never show confidence or assurance when expressing their ideas. This lack of confidence, coupled with poor intonation, results in limited discourse. Although the students demonstrate a willingness to participate, the observation sheet confirms that “fear of error” restricts the quality of their contributions.

Turn-taking and Disposition

In contrast to their linguistic deficiencies, the group shows an excellent disposition for working on oral activities and in pairs. Students always respect turn-taking and show a willingness to address everyday topics, which represents a solid foundation for introducing technological tools for practice.

Linguistic Proficiency and Instruction

Profound deficiencies were detected in the use of grammatical rules and syntax, which are considered very basic. The vocabulary used is limited and not always appropriate to the context, which hinders effective communication. This lack of linguistic resources generates frustration, as students have the intention to communicate but lack the grammatical tools to do so consistently.

Taking all these aspects into account, it is evident that students have a collaborative attitude and respect for speaking turns but face critical barriers that impede their oral performance. Insufficient fluency, the use of very basic grammatical structures, and reliance on isolated responses reflect a lack of linguistic tools that limit their spontaneous expression. Given this scenario, the implementation of AI is presented as a necessary solution to strengthen vocabulary and grammar in a safe environment, allowing students to overcome the insecurity detected and achieve more fluid and coherent communication.

Interview results

The interview with teachers at the Cinco de Mayo Educational Unit gave us a different perspective from the information obtained through observations. The following aspects stood out among the results.

Knowledge about Artificial Intelligence (AI)

Teachers express a balanced and thoughtful stance on the use of AI in education. In addition, they demonstrate prior knowledge of AI, acquired through institutional training, which shows a formal and responsible approach to this technological tool.

Influence of the implementation of Artificial Intelligence (AI) in the classroom

The teachers interviewed recognize AI as a pedagogical support resource that facilitates access to information and educational planning, although they emphasize that its implementation entails critical challenges such as student distraction and the difficulty of supervision, especially in virtual environments. In this case, the results suggest that AI does not replace the work of teachers but rather requires them to play an active role by establishing clear rules and providing constant support to prevent its inappropriate use. Finally, the teachers' perspective reaffirms the importance of human interaction as the central axis for ensuring that technology is used ethically and effectively, always geared toward the fulfillment of learning objectives.

Impact of Artificial Intelligence (AI) on students

The teachers interviewed agree that the use of AI has generated a disruptive change in the affective part of English language learning, highlighting mainly the mitigation of the “affective filter.” According to the testimonies, students show a greater willingness to produce language due to the absence of social judgments, since, when interacting with AI, the fear of ridicule or embarrassment for making mistakes in front of their peers disappears. Teachers observe that this digital “safe zone” allows students to take linguistic risks, supported by immediate and private feedback that corrects grammar and pronunciation in real time. As a result, teachers perceive AI not only as a technical tool, but also as emotional support that fosters autonomy and accelerates fluency, allowing students to arrive in the classroom with confidence strengthened by their private interactions with technology.

Proposed strategy for improving fluency in English

The main result of this research was a teaching strategy characterized by four closely related stages in which AI is used as a support tool.

Stage I. AI selection

For the selection of the AI tool, we sought to ensure that it met criteria of inclusivity, ease of access, and free availability to students. In addition, priority was given to platforms that allow for the practice of oral skills through simulated conversations, compatibility with mobile devices, immediate feedback, accessibility from different devices without deleting history or data, and an interface that does not require advanced technical knowledge. For this reason, among the tools evaluated, we suggest the use of Sesame AI, selected for its ability to interact conversationally at the students' level and its ease of use in educational contexts.

Stage II. Development of prompts for interaction with the AI

Prompts were designed to guide interaction with the AI so that students can practice the language in a guided and progressive manner based on the topic required by the class plan for that day. These prompts are focused on encouraging expression and fluency.

Some examples of prompts are:

- “Let's have a casual conversation. I want to talk about [Topic].”

- “I would like to practice using the [Present Perfect Continuous].”

- “Act as a [role, for instance, a hotel receptionist]. I am the customer, and I need [situation for instance, to check in or report a problem with a room].”

Stage III. Group organization

This strategy will be applied through activities developed in two modalities, individual and in pairs. In the individual modality, each student will interact directly with the AI to strengthen their confidence when talking about topics of their own interest when they manage to respond to the AI. Meanwhile, in the pair modality, students will share their experiences and results obtained with the AI, comparing responses and promoting collaborative learning.

Stage IV. Evaluation of the strategy

As a final stage, the strategy was evaluated through socialization workshops with teachers, who stated that the SESAME AI is “very motivating,” promotes self-correction and pair work, and facilitates language learning and interaction. They also considered that “the stages of the strategy are well structured and interrelated” and that the AI platform provides “emotional support,” which was considered relevant by the authors of this work, as student participation in a suitable psychological environment promotes the performance of both students and teachers.

Discussion

The findings derived from classroom observations and interviews with the two teachers show that communication and willingness to participate prevailed over linguistic accuracy, suggesting that students prioritized message transmission and effective interaction over formal language control, displaying vocabulary limitations, opting for the most common words, and unnatural intonation, reinforcing the findings of Bravo et al. (2024) who argue that traditional teaching often faces challenges due to a lack of tools that promote speech awareness and perceived shortcomings in learning that does not yet incorporate technologies as a support and corrective resource, thus causing the lexical and phonetic deficiencies previously observed.

Additionally, there is a similarity between the fluency observed in the students and the proposal by Alvarez Acevedo (2025) who validates that AI allows for a transition from fragmented communication to sequential oral production through simulated situations. However, from our observation, fluency occurred spontaneously due to the group's disposition, while the same author suggests that it should be guided by a structured methodology such as Task-Based Learning.

It is important to mention that, in the face of the irregularities presented, the students expressed a noticeable decrease in self-confidence, hesitations and doubts arose, confirming what the teachers had said about self-consciousness and reinforcing their idea of AI as a “disruptive” change by reducing the fear of social judgment, making them feel that they are in a “safety zone” where there is no ridicule or shame for making mistakes, coinciding with those who affirm that a “low affective filter” is the key to learning. Likewise, this thinking aligns with Acosta Rivera & Vaca, (2025) tolerance for error, arguing that these artificial tools provide dynamic interaction that favors a greater willingness to make mistakes, generating confidence and allowing student participation to continue even if grammar usage is irregular.

On the other hand, through the socialization workshop, teachers stated that they perceive SESAME AI as “very motivating” and “facilitating language learning and interaction”; “the stages of the strategy are well structured and related,” and that the AI platform is “emotional support” that allows students to take linguistic risks, because the tool allows for personalized and adaptive learning Chicaiza et al. (2023) where students practice privately and at their own pace. This idea is reinforced by Tramallino & Zeni (2024) who state that the AI's immediate guidance and correction of errors offers technical assistance that teachers, due to time constraints or the number of students, cannot always provide, thus resulting, according to Huerta et al. (2024) this results in a combination of emotional security and technical support, which increases motivation and active participation in the classroom.

Conclusions

After reviewing the literature and delving deeper into the theoretical foundations, it was determined that fluency in the English language is not only a linguistic process of verbal speed, but also a multidimensional phenomenon that depends closely on cognitive and affective factors. Through the theoretical review, it was identified that Artificial Intelligence (AI) and speech recognition tools act as an enhancer of cognitive fluency, facilitating the automation of grammatical and lexical structures.

In terms of methodology, a four-stage strategy was designed, with the use of Sesame AI as the central axis of this proposal, constituting an inclusive and accessible solution that allows learning to be detached from the physical space of the classroom. The suggestion of specific prompts, turn-taking, individual and team work, enhances guided interaction aligned with the real communication needs of high school students. Another benefit of the strategy is the immediate feedback that allows students to take linguistic risks often avoided in the traditional classroom, thus optimizing the fluency of their speech.

Finally, the proposal was shared and highly valued by the group of teachers, who described it as a capable and necessary tool for contemporary educational challenges. It is concluded that the effectiveness of the strategy lies in its ability to provide emotional support and virtual assistance, eliminating the fear of social judgment and ridicule. Thus, the integration of AI does not aim to replace the work of teachers, but rather to enhance it, giving students reinforced self-confidence and a consolidated foundation of fluency in an autonomous and private manner.

References

Acosta Rivera, L. G., & Barreto Vaca, C. del C. (2025). Impacto de la inteligencia artificial en el aprendizaje del inglés como lengua extranjera en estudiantes de bachillerato en Ecuador. Revista RedCA, 8(22), 140-164.

Alvarez Acevedo, A. de M. (2025). Uso de la inteligencia artificial generativa para mejorar la producción oral de los estudiantes de un curso de inglés universitario de nivel A1 [Tesis de maestría, Pontificia Universidad Católica del Perú]. Repositorio Digital de Tesis y Trabajos de Investigación PUCP. http://hdl.handle.net/20.500.12404/29955

Asensio Iglesias, P. (2024). El rol de la inteligencia artificial en los procesos de enseñanza y aprendizaje [Tesis de maestría, Universidad de Sevilla]. Depósito de Investigación Universidad de Sevilla. https://hdl.handle.net/11441/165187

Ayala Pazmiño, M., & Alvarado-Lucas, K. (2023). Integración de la inteligencia artificial en la educación del idioma inglés en Ecuador: Un camino para mejorar los resultados del aprendizaje. 593 Digital Publisher CEIT, 8(3-1), 221-234. https://doi.org/10.33386/593dp.2023.3-1.1862

Boers, F., Eyckmans, J., Kappel, J., Stengers, H., & Demecheleer, M. (2006). Formulaic sequences and perceived oral proficiency: Putting a lexical approach to the test. Language Teaching Research, 10(3), 245-261. https://doi.org/10.1191/1362168806lr195oa

Bravo, L. O. G., Pesantez, P. R. A., Fray, L. F. B., & León, A. E. M. (2024). La inteligencia artificial aplicada a la enseñanza del idioma inglés como lengua extranjera. Polo del Conocimiento, 9(8), 2791-2806. https://doi.org/10.23857/pc.v9i8.7854

Cha, J., Han, J., Yoo, H., & Oh, A. (2024). CHOP: Integrating ChatGPT into EFL oral presentation practice [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2407.07393

Chamorro Ortega, C. P., Bejarano Criollo, S. N., & Guano Merino, D. F. (2020). ¿Qué habilidad lingüística se hace más compleja enseñar en el aula de inglés como lengua extranjera? Domino de las Ciencias, 6(2), 308-325.

Chen, L., Chen, P., & Lin, Z. (2020). Artificial intelligence in education: A review. IEEE Access, 8, 75264-75278. https://doi.org/10.1109/ACCESS.2020.2988510

Chicaiza, R. M., Castillo, L. A. C., Ghose, G., Magayanes, I. E. C., & Fonseca, V. T. G. (2023). Aplicaciones de Chat GPT como inteligencia artificial para el aprendizaje de idioma inglés: Avances, desafíos y perspectivas futuras. LATAM Revista Latinoamericana de Ciencias Sociales y Humanidades, 4(2), 2610-2628. https://doi.org/10.56712/latam.v4i2.781

Deepgram. (2025). Yoodli AI: AI-powered speech and presentation coach. https://deepgram.com/ai-apps/yoodli-ai

EF Education First. (2025). Índice EF de dominio del inglés: Ecuador. Recuperado 10 de enero de 2026, de https://www.ef.com.ec/epi/regions/latin-america/ecuador/

Grasso, M. (2024). El desarrollo de la fluidez en la clase de lengua extranjera. Plurentes, (15), Artículo e079. https://doi.org/10.24215/18536212e079

Gupta, R. (2025). Artificial intelligence (AI): What it is, and its uses: A users guide [Manuscrito inédito]. https://doi.org/10.13140/RG.2.2.11218.26564

Huerta, S. G., Rosas, C. N. G., Hernández, Y. E., & Moreno, J. L. (2024). Uso de inteligencia artificial para desarrollar habilidades lingüísticas en el aprendizaje de inglés en bachillerato. Jóvenes en la Ciencia, 28, 1-5. https://doi.org/10.15174/jc.2024.4550

Jong, N. H. de, & Wempe, T. (2009). Praat script to detect syllable nuclei and measure speech rate automatically. Behavior Research Methods, 41(2), 385-390. https://doi.org/10.3758/BRM.41.2.385

Jordi. (2025, marzo 10). Sesame, el asistente de voz más natural. Blog de Inteligencia Artificial de Salesians Sarrià. https://ia.salesianssarria.com/sonido/sesame-evolucion-voz-conversacional-ia/

Lambert, C., Aubrey, S., & Leeming, P. (2021). Task preparation and second language speech production. TESOL Quarterly, 55(2), 331-365. https://doi.org/10.1002/tesq.598

López de Mántaras, R., & Brunet Crosa, P. (2023). ¿Qué es la inteligencia artificial? Papeles de Relaciones Ecosociales y Cambio Global, (164), 13-21.

López Regalado, O., Núñez Rojas, N., López Gil, Ó., & Sánchez Rodríguez, J. (2024). Análisis del uso de la inteligencia artificial en la educación universitaria: Una revisión sistemática. Pixel-Bit. Revista de Medios y Educación, (70), 97-122.

Martínez, R., Morales, J., & González, M. (2023). Inteligencia artificial en la educación. Revista Digital de Tecnologías Informáticas y Sistemas, 7(1), 100-106. https://doi.org/10.61530/redtis.vol7.n1.2023.136.100-106

Mendoza, F. M. C., & Torres, E. (2024). Impacto de las TIC en la enseñanza-aprendizaje: Caso de estudio en la carrera de Tecnología de la Información de la Universidad Técnica de Manabí. RIDE. Revista Iberoamericana para la Investigación y el Desarrollo Educativo, 15(29), Article e668. https://doi.org/10.23913/ride.v15i29.2099

Revelo Trujillo, J. E., & Ramirez Roman, P. A. (2024). Self-regulated cognitive strategies and ICT to enhance EFL speaking skills. Ibero-American Journal of Education & Society Research, 4(2), 1-12. https://doi.org/10.56183/iberoeds.v4i2.702

Romero, L., Gudayol-Ferré, E., & Padrós, F. (2019). Fluidez verbal, inteligencia y velocidad de procesamiento en adultos jóvenes con y sin actividad escolar: El impacto de la reserva cognitiva en adultos jóvenes. Revista Chilena de Neuropsicología, 14(2), 30-34. https://doi.org/10.5839/rcnp.2019.14.02.06

Romero, M. Á., Tiza, D. R. H., Murillo, J. P. M., Cervantez, D. O. O., & Ordóñez, G. I. (2023). Método mixto de investigación: Cuantitativo y cualitativo. En Instituto Universitario de Innovación Ciencia y Tecnología Inudi Perú. Instituto Universitario de Innovación Ciencia y Tecnología Inudi Perú. https://doi.org/10.35622/inudi.b.105

Salih, A. A., & Omar, L. I. (2024). Enhancing EFL learners' engagement in situational language skills through clustered digital materials. Frontiers in Education, 9, Article 1439104. https://doi.org/10.3389/feduc.2024.1439104

Segalowitz, N. (2005). Rethinking communicative language teaching: A focus on access to fluency. The Canadian Modern Language Review / La Revue canadienne des langues vivantes, 61(3), 325-353. https://doi.org/10.3138/cmlr.61.3.325

Segalowitz, N. (2010). Cognitive bases of second language fluency. Routledge. https://doi.org/10.4324/9780203851357

Sesame AI. (s.f.). Sesame AI: Natural human voice beyond the uncanny valley. https://sesameaivoice.com

Skehan, P. (2009). Modelling second language performance: Integrating complexity, accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510-532. https://doi.org/10.1093/applin/amp047

Tramallino, C. P., & Zeni, A. M. (2024). Avances y discusiones sobre el uso de inteligencia artificial (IA) en educación. Educación, 33(64), 29-54. https://doi.org/10.18800/educacion.202401.M002

Uribe, M. C., Cavallieri, I. R., & Rojas, G. A. F. (2024). Procesos que llevan a fomentar la fluidez en inglés como lengua extranjera (EFL) en un entorno universitario. Onomázein, (63), 110-125. https://doi.org/10.7764/onomazein.63.06

Conflict of interest

The authors declare that they have no conflicts of interest.

Declaration of responsibility of authorship

We, the authors of the aforementioned manuscript, DECLARE that we have directly contributed to its intellectual content, as well as to the genesis and analysis of its data; therefore, we are able to take public responsibility for it and accept that their names appear on the list of authors in the order indicated. Furthermore, we have complied with the ethical requirements of the aforementioned publication, having consulted the Declaration of Ethics and Publication Malpractice.

Bianca Hidali Pisco Tuarez, Diana Carolina Navarrete Parraga and Raisa Macias Sera: Literature review and article writing process.