Enhancing Digital Inclusion through AI-Based Yorùbá Language Localization: Challenges, Solutions, and Future Prospects

Abstract

In an increasingly digitized world, digital inclusion is critical for socioeconomic participation, yet linguistic barriers persist, particularly for indigenous language speakers. Yorùbá, a major West African language with over 40 million speakers, remains underrepresented in digital technologies, exacerbating digital exclusion and cultural erosion. This study explores the role of AI-based localization in bridging this gap, examining challenges, solutions, and future prospects for integrating Yorùbá into digital platforms. Using a mixed-methods approach, we analyse linguistic complexities like tonal variations, diacritic preservation, data scarcity, and technological limitations in existing Natural Language Processing (NLP) tools. Findings reveal moderate translation performance (BLEU: 32.4), tonal recognition challenges (WER: 21.7%), and high user satisfaction in text-to-speech applications (MOS: 4.1). Thematic analysis highlights demand for usability, cultural appropriateness, and linguistic empowerment. We propose AI-driven strategies, including multimodal systems, standardized linguistic resources, and ethical frameworks for cultural preservation. The study advocates for collaborative, policy-supported efforts to enhance digital inclusion, emphasizing the need for localized AI models and educational integration. By addressing these challenges, AI can empower Yorùbá speakers, preserve cultural heritage, and serve as a model for other marginalized languages.