The increasing presence of Artificial Intelligence (AI) in education has significantly influenced how English as a Foreign Language (EFL) learners develop spoken proficiency. This systematic review explores the most recent mobile platforms designed to support speaking development, applying the PRISMA methodology to ensure an accurate and thorough literature selection. Academic works were extracted from databases including Scopus, Web of Science, Google Scholar, and SciELO. The review focused on three guiding inquiries: (1) What AI-based mobile apps are used by English teachers? (2) How do such apps contribute to independent speaking practice? (3) What constraints are associated with their classroom integration? The analysis indicated that mobile applications positively impact oral fluency through features like real-time correction, individualized practice routines, and improved learner autonomy. These tools enhance flexibility in language acquisition, allowing learners to manage their practice schedules and receive targeted support. Nonetheless, implementation challenges such as technological inequality and algorithmic limitations were identified. Future exploration should address these concerns to maximize the pedagogical potential of AI-enhanced mobile learning. 8 ECOS DE LA ACADEMIA https://doi.org/10.53358/ecosacademia.v11i21 Vol.11 Núm.21 / enero a junio 2025 ISSN en línea: 2550-6889 Ethical concerns regarding data privacy and user consent also arise, as mobile applications often collect and store sensitive learner data (Kim & Choi, 2023). Therefore, while mobile applications offer substantial opportunities for enhancing oral fluency, their implementation must be accompanied by strategies to ensure equitable access, cultural sensitivity, data protection, and integration with human- mediated instructional support. Future research should continue to investigate how mobile technologies can best support autonomous speaking development while addressing these infrastructural and ethical challenges. Methodology Methodological Approach This review adopted a comprehensive and systematic approach to identify and evaluate scholarly works related to the use of mobile applications in fostering oral fluency among learners of English as a Foreign Language (EFL). The review process was grounded in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) protocol to maintain methodological precision. A targeted search was carried out across major academic databases, including Scopus, SciELO, Web of Science, and Google Scholar. The search strategy employed a combination of focused keywords and Boolean operators, using phrases such as: “mobile applications for EFL speaking skills,” “mobile apps in English language instruction,” “technology-supported speaking fluency,” “AI-powered speaking apps for EFL students,” “instant feedback in mobile language learning,” “self-regulated oral practice with mobile tools,” and “limitations of mobile technology in language teaching.” The methodological process comprised multiple stages: defining research aims and central questions; initial filtering of titles and abstracts to assess alignment with the review scope; application of inclusion and exclusion benchmarks during systematic searches; comprehensive review of full-text documents meeting eligibility standards; and data organization and synthesis to respond to the study’s guiding questions. To ensure data relevance and academic rigor, only peer-reviewed journal articles, conference 14 ECOS DE LA ACADEMIA https://doi.org/10.53358/ecosacademia.v11i21 Vol.11 Núm.21 / enero a junio 2025 ISSN en línea: 2550-6889 Despite their numerous benefits, the reviewed studies point to several persistent challenges that limit the effectiveness of mobile applications in enhancing oral fluency. Technological Limitations Although AI-powered tools have advanced significantly, they are not without flaws. Inaccuracies in speech recognition—particularly with diverse non-native accents— can lead to inconsistent or incorrect feedback, which may confuse learners or reinforce incorrect pronunciations (Sun et al., 2023). Additionally, some applications lack sensitivity to suprasegmental features such as rhythm and pitch variation, which are essential for natural-sounding speech. Ethical and Data Privacy Concerns Another underexplored area concerns the ethical implications of using AI-driven applications. Lee et al. (2023b) raise important questions about data privacy, noting that many apps collect and store users’ voice data without transparent consent processes. Furthermore, algorithmic bias remains a concern, as some apps perform better with standard varieties of English and may disadvantage speakers of less represented linguistic backgrounds. Teacher Preparedness and Professional Development Taylor and Garcia (2023) emphasize that many EFL educators lack the training needed to effectively incorporate mobile technologies into their instructional practice. Without adequate professional development, teachers may underutilize the pedagogical affordances of these tools or use them in ways that do not align with best practices in communicative language teaching. Mobile applications offer significant potential for enhancing oral fluency in learners of English as a Foreign Language (EFL), especially when they integrate AI-based feedback, engaging speaking tasks, and features that encourage learner autonomy. Nonetheless, their effectiveness is contingent upon thoughtful instructional integration, attention to ethical dimensions such as data privacy, and sufficient professional development for educators. These tools should not be viewed as replacements for traditional instruction but rather as complementary components within a blended learning environment—one that merges digital innovation with teacher-led guidance to support consistent, impactful language acquisition. 14

The proposed meta-analysis study examines the pedagogical role of artificial intelligence (AI)-based interventions in English as a Foreign Language (EFL) education through a synthesis of empirical evidence of 23 peer-reviewed experimental and quasi-experimental experiments published in 2019-2025. Scheduling a strict systematic review process in accordance with the PRISMA principles, the work investigates AI solutions, including chatbots, automated writing assessment platforms and virtual reality applications, assessing their effectiveness in relation to various demographic groups of learners, teaching environments, and language achievements. The effects sizes were calculated using the Comprehensive Meta-Analysis (CMA) software and g of Hedge on a random-effects model. The combined findings reported a statistically significant and large overall effect (g = 1.10, SE = 0.18, 95 % CI [0.75, 1.44]) that means that AI-ready pedagogies improve EFL learning outcomes significantly, especially regarding such aspects as the accuracy of writing, fluency in speaking, and motivation of the learners. In addition, other beneficial affective effects of using AI included the reduction of anxiety and motivation and enjoyment among the learners. Subgroup analyses also indicated that the kind of measurement tool had a pronounced moderating effect on effect size, with affective variables (e.g., motivation, engagement) having stronger increases than only quantitative variables (e.g., word count). Heterogeneity was high (I2 = 92.66) in order to highlight the role of contextual and methodological differences. The findings are part of the increasing literature on AI in language learning that provides empirical data on informing educational practices to guide further studies on the adoption of the concept of intelligent technologies in EFL teaching.

Over the past decade, studies in EMI contexts describe a range of music-based interventions for EFL learners that generally fall into several categories. Many interventions rely on song listening as the core activity. Some studies use song listening alone, while others combine it with activities such as lyric analysis, dictation, gap-filling, or group discussion. Digital platforms-via Spotify, YouTube, LyricsTraining, and other web-based tools-have also been deployed to deliver authentic music experiences and provide interactive feedback. A smaller set of studies reports the use of music cloze exercises, jazz chants, pop music selections, children's songs, humorous songs, and even singing paired with body movement or drama to enhance engagement and fluency. Quantitative reports detail meaningful improvements. For example, one study documented an increase in mean engagement scores from 3.06 to 7 following a 10-week song listening intervention, while another showed test scores rising from 65% to 82% when song listening was coupled with lyric analysis. Other studies note enhancements in reduced form recognition, listening comprehension, and overall motivation. Together, these findings illustrate that music-based interventions-whether grounded in traditional song listening or enhanced by digital and interactive components-are associated with increased listening engagement and measurable gains in fluency.