ADVANCED FEATURE EXTRACTION AND ITS IMPLEMENTATION IN SPEECH RECOGNITION SYSTEM

Abstract

IN SPEECH RECOGNITION SYSTEM” deals with the in depth study of Feature Extraction module. It encompasses various feature extraction algorithms which would be implemented and optimized. Different parameters would be extracted both in time domain and frequency domain. Speech Processing is an upcoming field. A lot of research is conducted in this field. Today people are interested in hand free systems. A password can be hacked but using voice as a password can not be hacked. We want to make the use of mouse obsolete. As we speak ‘my computer’, it should be double clicked or in general the respective application should be opened. The project is mainly concerned with the advanced feature extraction and implementation of those features in the speech recognition system. Various feature extraction algorithms have been implemented in the project. They are then compared with each other with respect to their performance. Graphs have been plotted to represent their respective performance in terms of their recognition rates. The project is designed following a simplistic yet an effective approach so that the user has a hassle free working experience. This project includes text fields for the input word and recognized word and the list of various levels of noise which need to be added. I have used its application in design of voice automated robot. Voice based robotic control is an interesting project, mainly used for industrial and surveillance applications. It gives exact concept of controlling a robot by a voice instruction. 5 Robot is capable of understanding and synthesizing human speech for communication. A voice recognition unit built around a high speed processor that ensures various operations of the system to be performed by voice command. A few of commands recommended for operation are listed as: START, STOP, FORWARD, REVERSE, RIGHT, LEFT, SLOW, FAST,OK, UP, DOWN, CLOCK, ANTICLOCK, CLOSE,OPEN. After the design of voice automated mobile robot I have designed the gesture automated mobile robot. Robot is capable of understanding and synthesizing gesture for communication.