We present a computer-assisted language learning system thatautomatically evaluates the pronunciation and fluency of spokenMalay and Tamil. Our system consists of a server and a user-facing Android application, where the server is responsible forspeech-to-text alignment as well as pronunciation and fluencyscoring. We describe our system architecture and discuss thetechnical challenges associated with low resource languages. Tothe best of our knowledge, this work is the first pronunciationand fluency scoring system for Malay and Tamil.
This work is supported by the Agency for Science, Technology and Research (A*STAR) under its A*ccelerate Gap Fund (ETPL/18-GAP033-R20A).