Speech recognition software converts words spoken into a microphone into machine-readable format. Speech recognition products can provide an appropriate computer input method for some individuals with a wide range of disabilities.
Speech recognition technology requires that the user have a relatively consistent speaking voice and moderately good reading skills in order to recognize errors and correct the program's text output. Voice and breath stamina should be considered when this is evaluated as a potential input method. The accuracy of many speech recognition systems should improve over time with consistent and accurate correction by the user.
For more information consult Working Together: Computers and People with Mobility Impairments, and Working Together: People with Disabilities and Computer Technology.