2 Finite States Techniques for Dialogue Processing

A common technique in dialogue processing is the use of a kind of FSA to specify the dialogues that a sysem can handle. The Saarbrücken Speaking Elevator mentioned earlier is based on such a FSA. We will now introduce FSA techniques for dialogue processing and then discuss the elevator's underlying FSA.

• 2.1 FSA-based Dialogue Systems
  • 2.1.1 Processing dialogue with finite state systems
  • 2.1.2 A simple FSA example

• 2.2 Extending FSA
  • 2.2.1 Why Extending FSA?
  • 2.2.2 Grounding Extension 1
  • 2.2.3 Grounding Extension 2
  • 2.2.4 Grounding Extension 3
  • 2.2.5 And what about Repair?

• 2.3 An In-depth Look at the Speaking Elevator
  • 2.3.1 The Saarbrücken CL department's Speaking Elevator
  • 2.3.2 How does the dialogue progress?
  • 2.3.3 An additional memory
  • 2.3.4 How long does the system wait for a response?
  • 2.3.5 How does the elevator deal with unrecognised utterances or inconsistent input?
  • 2.3.6 And what happens if the elevator does understand the input?
  • 2.3.7 How is the user input confirmed?
  • 2.3.8 During the trip...
  • 2.3.9 What happens when there is ambiguous input?
  • 2.3.10 Speaking elevator reference

• 2.4 Dialogue Characteristic and Extension of the Elevator
  • 2.4.1 Turn-taking
  • 2.4.2 Adjacency pairs and insertions
  • 2.4.3 Grounding
  • 2.4.4 Dialogue context
  • 2.4.5 Ellipsis
  • 2.4.6 Reference resolution
  • 2.4.7 Mixed initiative

• 2.5 Summary and Outlook
  • 2.5.1 Advantages and disadvantages
  • 2.5.2 The CSLU tool
  • 2.5.3 Beyond Finite State Techniques