Spoken dialog system

A spoken dialog system is a computer system able to converse with a human with voice. It has two essential components that do not exist in a text dialog system: a speech recognizer and a text-to-speech module.

Components

• Speech recognizer
• Text-to-speech
• Response generator, which uses techniques beyond natural language generation. A response generator has to take domain information into account. Different wording of saying the same thing to the user will yield very different results. In call center application, response design is a very important art.
• dialog manager
• knowledge base
• Natural language understanding

Types of systems

Spoken dialog systems vary in their complexity. In general, the informational dialog system is the most simple one, and it gradually goes up to the most complex type, the entertainment dialog system.

• Informational: Examples include stock quotes, credit card and bank information. This also include Google 411 and Tellme directory assistance.
• Transactional: Examples include magazine subscription, and bank transaction.
• Diagnostic: Technical support.
• Educational/Tutoring: Tutoring system for physics and language learning.
• Entertainment and chatting

References

• Gabriel Skantze, Error Handling in Spoken Dialogue Systems, 2007: chapter 2, Spoken dialogue systems.
• Michael F. McTear, Spoken Dialogue Technology