Man-machine speech 'dialogue acts'.

Area of Science:

• Human-Computer Interaction
• Speech Technology
• Computational Linguistics

Background:

• The increasing prevalence of speech as a human-machine communication interface necessitates structured dialogue design.
• Effective man-machine interaction relies on understanding and organizing conversational elements.

Purpose of the Study:

• To propose a preliminary classification of man-machine speech dialogues.
• To outline essential features for interactive information services.
• To suggest a hierarchy for data entry modes in speech-based systems.

Main Methods:

• Classification of dialogues into 'dialogue acts' and 'subdialogues'.
• Identification of general features for interactive information services.
• Proposal of a data entry hierarchy.

Main Results:

• A preliminary classification system for speech dialogues is presented.
• Key features for interactive speech-based information services are identified.
• A structured approach to data entry modes is proposed.

Conclusions:

• Structured dialogue acts and subdialogues are crucial for effective man-machine speech communication.
• The proposed framework supports the development of advanced interactive speech information services.
• Further research can build upon this classification for more sophisticated dialogue management.