The Automatic Reading Tutor, an open-source web application, has been developed to address significant literacy challenges in South Africa. Tailored primarily for English speakers, this tool uses advanced speech recognition technology to enhance reading proficiency among adult and older child learners.
Despite not pinpointing specific mispronounced words, the application successfully indicates reading inaccuracies and provides immediate, automated feedback on overall reading performance. Users receive visual cues and corrective assistance in real-time, which helps in fostering self-directed learning and improvement.
Additionally, the tool collects and presents data on user progress, including reading speed and accuracy, although its effectiveness can vary with accent differences due to its reliance on standard speech recognition models. This limitation notwithstanding, the Automatic Reading Tutor represents a significant step forward in educational technology, offering substantial support to learners striving to improve their reading skills in challenging environments.
https://gitlab.cs.uct.ac.za/rmkyas002/read-frontend-application.git