Enhancement of speech communication between human and machine

Speech is a most natural and efficient way to exchange information for humanbeings. To create this information exchange between machine and humanbeing and to make a real ?intelligent computer,? it is important that themachine can ?hear?, ?understand,? and ?act upon? spoken information, andalso ?speak? to complete the information exchange. Therefore, ?speechrecognition? is essential for a computer to reach the goal of natural human-computer communication. Speech Recognition is a fascinating field spanningseveral areas of computer science and mathematics. Reliable speechrecognition is a hard problem, requiring a combination of many techniques;however modern methods have been able to achieve an impressive degree ofaccuracy but the improved algorithms still have some problems.This project attempts to examine one of those techniques, and to apply them tobuild a sample voice recognition system for Turkish. The goal of this thesis isto be able to recognize isolated words, spoken individually. In this work, thespeech data is loaded in to the computer using data acquisition card. This datais processed by software prepared in MATLAB. Linear Predictive Coding andDynamic Time Warping method is used in the program algorithm.Keywords: signal processing, speech recognition