DESIGN AND IMPLEMENTATION OF A SPEECH RECOGNITION AND SYNTHESIS SYSTEM

Undergraduate

ABSTRACT

As a new convenient means of human-machine interaction, voice command is widely applied to many portable embed speech products. The ultimate aim of speech recognition is to make machine understand natural language. It is of great significance not only in practical application but scientific research. Speech recognition basically means talking to a computer, having it recognize what we are saying, and lastly, doing this in real time. The research on speech recognition technology mainly concentrates on two aspects. One is the software running on computer, the other is embedded systems.


INTRODUCTION

Kimberlee (2009) says, it is a well-known trend in computer science to adapt computers to suit their users and not vice versa. The exploitation of more human senses in human- machine interaction than just sight can be proved to be more efficient and natural.

Kemble (2009); one way of enhancing human-machine interaction is to supply an additional response channel by means of speech recognition and synthesis. Simply formulated, speech recognition can be regarded as the automated ‘understanding’ of spoken words. In the third section we describe speech recognition in more detail.

According to Dmitry (2008), Speech recognition and synthesis allows interfaces to be built that permit users to interact with computers using natural language. By doing so, it widens the bandwidth of the communication channel that exists between user and computer. However, Dmitry (2008), also point out that speech recognition is not a panacea. Some input/output tasks (e.g., numerical data entry) are such that using SRS could decrease the interface performance. Dmitry (2008), furthermore says, natural language can be ambiguous. It is of great importance, therefore, not to implement speech recognition in each and every application, but to use it carefully. For instance, in critical applications such as nuclear plant control, key-entered input would be more appropriate.

Jeff  (2002), prompt that it is also unwise to conclude that ‘plugging in’ a speech recognition device offers the best way of communication, even though the situation seems suitable because using speech recognition draws on aspects in the domains of linguistics, cognitive psychology, etc. Jeff (2002), continue to It is not enough just to present the user with the opportunity to use a speech recognition program without taking human factors into account. Not supporting the user in an appropriate way during the interaction process will diminish the benefits of speech recognition. If, on the other hand, human factors are taken into account, a natural communication channel can be created Jeff (2002).  


STATEMENT OF THE PROBLEMS

Computers in this new age are made user-friendly for easy operation, despise that, some users’ find the operation difficult especially handicaps individuals. Some of this difficulties are;

Ø Difference in the arrangement of keyboard

Ø Improper legibility of text readability

Ø  Computer operations are not flexible

Ø Keyboard and mouse operations are old operational technology

Ø Time consuming in terms of input provision.

AIM AND OBJECTIVES OF THE STUDY

The aim of this study is to build an accurate and efficient speech recognition and synthesis system which is embedded with the capability of understanding human speech done in English language. 

Objectives:

Ø To simplify computer operation.

Ø To enable remote control of computer system by the use of speech synthesis system or recognition system.

Ø To enable interaction between user and the computer system.

Ø Keyless computer operation and authentication