English pronunciation verification system for Apple iOS

• Main Author: Ng, Kelvin Sio Joo.
• Other Authors: Soon Ing Yann
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/52464
• Institution: Nanyang Technological University

Each culture has pronunciation patterns and lilts unique to that culture and these unique elocution styles, while accepted and taken for granted in that culture, can render a word unintelligible in another culture. This calls for an easily accessible application that can quickly give the users affirmation about their pronunciation. This project uses PocketSphinx, a speech recognition library developed by Carnegie Mellon University (CMU) to implement the Speech Recognition and Speech Synthesization of predetermined commonly-used phrases and idioms within an application on the iPhone platform without connecting to the servers online for recognition processing. This application receives speech inputs from the users, analyses the inputs, searches and matches the inputs with the creation of dictionary and Language Model (LM) available and returns the best match by probability. Developing this speaker independent application enables users to verify their pronunciation of the phrases and provides users fast, reliable and accurate pronunciation, definition of phrases and recognition during execution.