First, to operate a computer through voice, the user must learn how to dictate in a word-by-word manner known as "discrete speech." In other words, the computer cannot recognize individual words if they are spoken the way people usually speak--in fluent sentences or "continuous speech." Next, the user must "teach" the system to recognize his or her voice through a combination of training and usage. We all pronounce individual words in different ways, and voice recognition software cannot simply recognize everyone's voice right off the bat.
As the user speaks to the system, the software creates a user-specific voice file that contains a lot of information about his or her voice qualities and pronunciations. The system uses this information to make its best guess at what each word is as it is dictated.
The process of "familiarizing" the voice recognition software with an individual voice takes time. When a user takes the time to properly train and use the voice recognition system, which creates a strong and accurate voice file, the system will supply the correct word most of the time. However, the system will never achieve a 100% accuracy rate in all situations. Sometimes the computer just doesn't get it right and suggests the wrong word. The user must then stop and correct the system.