Bengali Segmented automated speech recognition

Abstract

Speech recognition and understanding of spontaneous speech have been an elusive goal of research since 1970. For understanding speech human not only consider for information passed to the ears but also judge the information by the context of the information. That’s why human can easily understand the spoken language convey to them even in noisy environment. Recognizing speech by machine is so difficult for the dynamic characteristics of spoken languages. People used different approaches for automated speech recognition system. For recognizing speech people always prefer English as most of the research and implemented for them. So I am intended to have my research on Speech Recognition system but preferably in our mother tongue –Bengali. It is an area where a lot to contribute for our language to establish in computer field. The contribution of this thesis is to show how to build a speech recognizer using HTK toolkit which can recognize Bengali words. Bengali speech recognizer is built by training the HTK toolkit and can recognize any word in the dictionary. After acoustic analysis of speech signal waves the words are recognized. Technically this thesis presents training the toolkit and builds a segmented speech recognizer of Bengali. Finally the thesis contains the training procedure of the toolkit, how people can build a recognizer with the HTK toolkit.