First of all, the computer picks up speech via a microphone which gives an acoustic, analogic signal.
<br>The sound card converts this analogic symbol to a digital symbol.
<br>
The speech engine then converts the digital symbols into phonemes and words that can be compared against a grammar or dictionary.<br>
Speech-aware application processes the words and commands as though they had been entered by the keyboard or the mouse, as is the case for dictation applications.
Dictations, language learning, researching and retrieving data, commanding browsers, games and all sorts of multimedia and non-multimedia applications such as industrial, medical, financial and legal data processing.
There are no limitations to speech-aware applictions.
Turn to ½ WhatÆs it for ╗ for some convincing examples.
Some speech-aware applications are more interactive than others and may talk back to you! In this case they are equiped with a speech synthesis engine, or a text to speech function, which does exactly the opposite of the speech engine: it converts words into phonetic and prosodic symbols. These then become a digital audio stream which you hear through your loudspeakers.
For a more natural feedback, real speech can be pre-recorded in sound files that are automatically played back to the user according to the program commands.
Continuous speech and voice commands are already permanent features in some word processors and have yet to be installed in other applications to replace mouse and keyboard inputting. This is the mission of Theorom Productions.
We have forgotten how long the mouse took to be accepted, but now no set-up is conceivable without it. Speech is going the same route
As our knowledge of natural language evolves, voice technology will simulate human speech and human thought to such a point that talking back to your computer will become reality.
