Duolingo evolution: From automation to Artificial Intelligence

This paper explores the transformative journey of the Duolingo language learning application from automation to Artificial Intelligence (AI) integration. Through secondary data analysis, it traces the historical trajectory of Computer Assisted Language Learning (CALL) in conjunction with the pedagogical and technological advancements. By examining Duolingo as a case study, the paper investigates the motivations behind its shift to AI and the impact of evolving educational methods on this transition. Key findings highlight that Duolingo’s adoption of AI, particularly through Automatic Speech Recognition (ASR), has significantly enhanced language learning by offering personalized, interactive experiences, and immediate feedback. The study delves into the five key stages of automated language assessment, and details Duolingo’s migration from Phyton to Scala, addressing the challenges and strategic decisions that improved scalability, performance, and stability. Despite obstacles, like documentation gaps and compatibility issues, Duolingo’s technological advancements have been well-received by users, indicating high usability and user satisfaction. Finally, Duolingo’s evolution mirrors broader trends in technology-driven language education, emphasizing the importance of innovation in meeting the evolving needs of learners in the digital age.