This preview shows page 1. Sign up to view the full content.
Unformatted text preview: ms are much more difficult to build than speakerdependent systems.
According to another classification mechanism, voice recognition systems are of the
following two types:
1. Single word recognition systems. These systems are designed to recognize only a
single spoken word, such as YES, NO, MOVE, STOP, at a time. Speaker-independent
systems are mostly of this type.
Continuous speech recognition systems. These systems are designed to recognize
spoken sentences such as MOVE TO THE NEXT BLOCK. Since recognizing
continuously spoken sentences is more difficult than recognizing words, continuous
speech recognition systems are speaker-dependent.
Although in its infancy, voice recognition systems are already being used for a wide
range of applications. Some of its typical applications are as follows: 1. For inputting data to a computer system by a person in situations where his/her hands
are busy, or his/her eyes must be fixed on a measuring instrument or some other object
For example, a quality-control person who must use his/her hands to detect defects can
call out the identification number of the sample found defective as they are detected;
persons handling baggage at airports can simply speak out the three-letter destination
identifier for the luggage to be automatically routed to the appropriate conveyor system;
and doctors in an operation room can request certain information about a patient while
2. For authentication of a user by a computer system based on voice input.
3. For limited use of computers by individuals with physical disabilities.
In addition to making input of data easier, voice recognition systems also provide
tremendous freedom of movement to the operator because the operator is free to stand up
and move around while inputting voice data to the system.
A vision-input system allows a computer to accept input just by seeing an object. The
input data in this case is normally an object's shape and features in the form of an image.
The idea is to simulate the capability of a human vision system in a limited sense.
View Full Document
- Spring '14