The fundamental mode of communication among humans is speech and in the case of
machine-human interface; verbal language has been believed as the natural method. When
communication with machines is carried out, it is very difficult and slow-moving in
magnitude when realized via keyboards, mouses and other devices. Thus, speech feed-in is an
important constituent to making this communication easily accessible. Also, humans see speech as a
great source of information. Therefore, persons who are not literate or have vague about computers
can easily access computers by employing speech instructions. Even people with some physical
disability who are not able to type or click with their hands can use their speech to operate the
computer. Even people who are proficient in operating computers can speed up data entry, sending
emails and other documents using the speech input methods. Furthermore, this mode of
operation possesses many advantages. For example, while driving, the hands of the driver are
busy steering the driving wheel and he cannot type on his mobile. In such a case speech
is a good input option. GPS (Global Positioning System) is an example of a speech-based
system being used. Another example is speech-enabled dialing, where the user can just ask the
device to call a particular person, without dialing his number.
The common and efficient means of communication among humans is through 'speech'. To process speech
means to extract useful information from it, processing includes the implementation of electric
signals on the acoustic pressure waves collected from human vocalization and applying
mathematical analysis to it. The field of processing speech involves the natural operation
of analysing speech, coding, augmentation, synthesis, and recognition. Analysis of speech is
the study of its creation mechanism to make a mathematical model of physical phenomena.
Speech coding aims to keep information about specific speech parameters for later retrieval. The
method of refining precision and quality of speech which is noisy utilises various algorithms is
recognized as speech enhancement [2]. Producing artificial human speech using coded information is
known as the synthesis of speech. The method of inverse synthesis is the capability of a program or
machine to classify the linguistic contents mixed up in the speech signal.
