![]() | Only 14 pages are availabe for public view |
Abstract Speech recognition is the process of automatically extracting and determining linguistic information conveyed by a speech wave using computers or electronic circuits. This area is an extremely important and has a tremendous potential for widespread use in the field of man-machine communication by voice. Within the general area of man-machine communication by voice, there are three major areas (modes of communication) to be recognized. These areas include: l)voice response systems, 2)speaker recognition systems, and 3)speech recognition systems. Voice response systems are designed to respond to a request for information using spoken messages. Thus voice response systems communicate by voice in one direction only, i.e from the machine to man. On the other hand, areas 2 and 3 in the above list deal with systems in which communication is by voice from man to machine. For speaker recognition systems the task of the system is to either verify a speakers identity or to identify the speaker from some known ensemble. Thus the speaker recognition area is itself broken down into the sub-areas of speaker verification and speaker identification. |