Multi-Purpose Speech Recognition and Speech Synthesis System

A multi-purpose speech recognition and speech synthesis system which includes applications that employ Speech Recognition (speech-to-text) and Speech Synthesis (text-to-speech) technologies. The input of the system is a speech signal, and it can also be typed text or graphical triggers. The output can be a signal signal or a text, along with graphical user interface forms. The applications are dictionary, calculator, search engine, movie guide, movie search, news reader, address book, image gallery and a weather forecaster. This book includes explanations of Speech Recognition, Speech Synthesis, and the functionality of the system and its applications. It also contains an explanation of how a special hand-gesture-controlled cursor and its hardware were developed. The system was developed to be a graduation project of Hussein Mohsen, Karim Jahed, and Marwan Fawaz at the Lebanese American University (LAU), Beirut, Lebanon, for Spring 2011 semester. The project was supervised by Dr. Nashat Mansour, Professor of Computer Science and Assistant Dean of School of Arts and Sciences at LAU-Beirut.