Kutti AI: A Voice-First, Offline-Capable Learning Companion with Real-Time Struggle Detection for Visually-Impaired Children

Kutti AI addresses a persistent equity gap in educational technology: nearly all edtech assumes a visual interface, excluding an estimated 1.4 million blind children worldwide. The system inverts this assumption entirely, making spoken conversation the primary and sufficient learning modality — children hear curriculum content, answer aloud, and receive spoken feedback with no visual dependency, positioning it within the Special Education and accessibility strand of Adaptive Learning research.

Three engineering contributions make this practical on commodity mobile hardware. First, a multi-signal struggle-detection engine fuses response latency, wrong-attempt counts, and keyword-based hesitation cues to decide in real time when to offer hints or simplify questions — a lightweight alternative to the learner-modeling machinery of full Intelligent Tutoring. Second, a cross-language answer-matching pipeline (translation/transliteration, Levenshtein fuzzy matching, text normalization) ensures children are not penalized for code-switching or pronunciation variation, an important fairness property for multilingual learners and a concrete instance of Equity-aware design. Third, an offline-first on-device ASR pipeline removes the connectivity requirement, extending Personalized Learning to low-resource settings where cloud-dependent tutors fail.

The paper is a systems contribution rather than an efficacy study — no learning-gains evaluation is reported — so claims about pedagogical impact should be treated as design hypotheses pending classroom trials. Nonetheless it is a rare example of Student Experience research that centers disabled learners from the outset rather than retrofitting accessibility.

Connected Concepts

Personalized Learning

Student Experience

Connected Articles

Elevate GenAI Virtual Tutors — ELEVATE: Designing Human-Centered GenAI Virtual Tutors for Scalable and Inclusive Education

AI LMS Middle School Longitudinal — AI-Integrated Learning Management System for Middle School: A Longitudinal Study of Learning Outcomes

Tactile Statistical Graphs Accessibility — Touching and Feeling the Data: A Reusable Software Pipeline for Tactile Statistical Graphs in Accessible Education

Pattern Kc Programming Recommendation — Automated Recommendation of Programming Learning Content Using Pattern-based Knowledge Components

AI Powered Personalized Learning Elementary Fractions 2026 — Exploring Fraction Comprehension and Interest in Elementary Education Through AI-Powered Personalized Learning

Kt4eqg Personalized Question Generation — KT4EQG: Personalized Exercise Question Generation via Knowledge Tracing

Citation

Kadharmoideen Fadurudeen (2026). Kutti AI: A Voice-First, Offline-Capable Learning Companion with Real-Time Struggle Detection for Visually-Impaired Children. arXiv:2607.22377. arXiv preprint.