Spoken Computer

From: Toward Spoken Human–Computer Tutorial Dialogues

Oral discourse is the primary form of human–human communication, hence, computer interfaces that communicate via unstructured spoken dialogues will presumably provide a more efficient, meaningful, and naturalistic interaction experience.

The ability to have a natural language spoken conversation with a computer is a technological feat that has been long desired, but the achievements have slowly accumulated and have arguably been modest. Developing artificial spoken dialogues in natural language involves an integration of systems that combine automatic speech recognition (ASR) for speech-to-text translation with natural language processing to link the recognized text to specific computer actions. Such systems mark a significant departure from the currently dominant window, icon, menu, and pointing device (WIMP) style of interaction.

The major motivating factor behind redefining the interaction paradigm from WIMP to spoken dialogues is to narrow the communicative barrier between computers and humans. Humans primarily communicate through speech and a host of nonverbal cues, such as facial expressions, posture, and gesture, rather than through typing and clicking. Computer systems that are able to recognize and respond to these communication channels will presumably provide a more efficient, meaningful, and naturalistic interaction experience. The ability to communicate with a computer though natural speech would represent a significant advancement toward cracking the barrier between the highly expressive human and the socially challenged computer.