Speech-recognition systems allow users to interact with them by using spoken commands
Speech-output systems use artificially generated speech
e.g. written-text-to-speech systems for the blind
Natural-language systems enable users to type in questions and give text-based responses