AT and Learning Disabilities

Speech Recognition

Speech recognition systems allow a person to operate a computer by speaking to it. In combination with a word processor, the user dictates to the system through a microphone. The user can speak either with pauses between words (discrete speech) or in a normal talking manner (continuous speech). The discrete product, although slower, may be the better choice for students with learning disabilities because errors can be identified as they occur. With continuous speech, making corrections after the fact requires good reading skills. Speech recognition technology requires that the user have moderately good reading comprehension to correct the program's text output. If the system incorrectly recognizes a word, the user can choose the correct word from a list of similar sounding words displayed on the screen. The more the system is used, the better it becomes at recognizing the user’s spoken language. Speech recognition systems may be most helpful to students whose oral language abilities are stronger than their written language abilities. It entirely eliminates the act of spelling, as well as keyboarding or handwriting, allowing a student to focus entirely on sentence structure, rhetoric, and critical analysis.

Sophisticated systems allow a person to dictate from 40-70 words a minute. The systems eventually learn the phonetic characteristics of each person’s voice. The more the system is used, the better it is able to understand the user. The success of these systems depends on the ability of the person to “train” the computer, to distinguish between a mistake in “hearing” by the computer or in “talking” by the user. The training process is intense and new users can become frustrated before getting to the point where voice input is successful. It takes patience and, for many students, consistent support from others, even several one-on-one sessions with an experienced instructor. Often a higher quality microphone works better than the one that comes with the software and it is important that the same microphone be used consistently.

Examples:

back to top

SCATP Home