Jehsh Philemon Mulwahali

Yekiva is an offline-first, voice-guided literacy application for children and adults who are beginning to learn how to read. It solves the problem of text-dependent learning software by allowing learners to use the application even before they can read menus or instructions. Yekiva supports French, Swahili, and Lingala, and guides learners through letters, sounds, syllables, words, and sentences using voice, visual cues, repetition, and interactive quizzes. Each letter is taught in four forms: uppercase and lowercase printed, plus uppercase and lowercase handwritten. Quiz choices are limited to learned content, while final games introduce new words to test transfer. Yekiva can also turn a photo of an everyday object into a lesson, even when the image contains no text. It identifies the object, names it in the selected language, describes it, and generates an interactive activity. A dedicated voice conversation feature lets learners speak, receive an AI-generated response, and hear that response aloud. The project uses Meta models for speech transcription, image understanding, conversation, and lesson generation. Yekiva makes literacy learning more accessible, multilingual, practical, and inclusive for learners who cannot read yet.