Page 45 - DCAP303_MULTIMEDIA_SYSTEMS

Unit 3: Sound

Levels notes
If the equipment or software you are using for your audio has level meters, you basically want
to have the meters just brushing near the red without going into them. This is the 0 dB marker.
Going over this will cause clipping and distortion of your recording.

What good is the video you shot if you cannot understand what the subject is saying? You have
probably seen a few home videos with either low fidelity audio which sounded tinny or noisy
or which was too low in volume to make out. This is not the videotape’s fault. It is the camera’s
built in microphone. As noted above, higher quality audio compresses more cleanly. An added
side benefit is that the more you can compress your audio, the more space you have for video,
resulting in higher quality video!

3.8 voice recognition and response

Voice recognition and voice response promise to be the easiest method of providing a user
interface for data entry and conversational computing, since speech is the easiest, most
natural means of human communication. Voice input and output of data have now become
technologically and economically feasible for a variety of applications.
Voice recognition systems analyze and classify speech or vocal tract patterns and convert
them into digital codes for entry into a computer system. Most voice recognition systems
require “training” the computer to recognize a limited vocabulary of standard words for
each user. Operators train the system to recognize their voices by repeating each word in the
vocabulary about 10 times. Trained systems regularly achieve a 99% plus word recognition
rate. Speaker-independent voice recognition systems, which allow a computer to understand
a voice it has never heard before, are used in a limited number of applications.
Voice recognition devices are used in work situations where operators need to perform data entry
without using their hands to key in data or instructions. For example, voice recognition systems
are being used by manufacturers for the inspection, inventory, and quality control of a variety
of products, and by airlines and parcel delivery companies for voice-directed sorting of baggage
and parcels. Voice recognition is also available for some microcomputer software packages for
voice input of data and commands. However, voice input is expected to become very popular
for most word processing applications in the next few years.
Voice response devices range from mainframe audio-response units to voice-messaging
minicomputers to speech synthesizer microprocessors. Speech microprocessors can be found
in toys, calculators, appliances, automobiles, and a variety of other consumer, commercial, and
industrial products. Voice-messaging minicomputer and mainframe audio response units use
voice-response software to verbally guide an operator through the steps of a task in many kinds
of activities. They may also allow computers to respond to verbal and touch-tone input over the
telephone. Examples of applications include computerized telephone call switching, telemarketing
surveys, bank pay-by-phone bill-paying services, stock quotations services, university registration
systems, and customer credit and account balance inquiries.

History of Computer sound

n the 1940s, the transistor allowed the construction of computers that were sophisticated
for their time. A computer is unique among machines in that its function is not inherent in
Iits structure, but rather determined by a set of instructions entered into it. It made sense,
therefore, that someone would develop a program that would allow computers not only to
store and transmit music, but also to compose and perform it. Performance is facilitated by
Contd...

LoveLy professionaL University 39