Spoken Dialog Systems and

advertisement
Spoken Dialog Systems and Voice XML
Lecturer: Prof. Esther Levin
Description
 Spoken dialogue systems enable users to interact with
computer systems via natural and intelligent dialogues, as
they would with human agents.
 Development of such systems requires a wide range of
speech and language technologies, including
 automatic speech recognition (ASR), to convert audio signals
of human speech into text strings,
 natural language and dialogue processing (NLP), to determine
the meanings and intentions of the recognized utterances and
to generate a cooperative response to them,
 and text-to-speech synthesis (TTS), to convert the system
utterance into actual speech output.
VoiceXML
 VoiceXML is the HTML of the voice web, the open standard
markup language for voice applications.
 HTML assumes a graphical web browser with display,
keyboard, and mouse,
 VoiceXML assumes a voice browser with audio output,
audio input, and keypad input. Audio input is handled by
the voice browser's speech recognizer. Audio output
consists both of recordings and speech synthesized by the
voice browser's text-to-speech system.
 VoiceXML takes advantage of several trends:
 The growth of the World-Wide Web and of its capabilities.
 Improvements in computer-based speech recognition and
text-to-speech synthesis.
 The spread of the WWW beyond the desktop computer
Project Scope
 We will be designing, developing, testing and deploying spoken
dialog system for a variety of applications.
 Upon the successful completion of this project a student should be
able to:
 understand the main functional components of a typical spoken
language processing system;
 have a detailed knowledge of the basic elements of spoken
language technology, such as patterns recognition, Hidden
Markov Models, and speech recognition
 have practical experience of speech recognition technologies
and of spoken dialogue system development using VoiceXML;
 appreciate current research issues in spoken language
technology and be aware of its commercial applications
Logistics
 In this project-based course, students are grouped into teams to
work on projects involved with design, implementation and testing
of spoken dialog systems.
 The capstone course will last two semesters.
 In the first semester, we will study key technologies involved in
this multi-disciplinary field.
 The second semester will focus on implementation of exciting
real-world dialog systems using the Voice XML platform.
 There will be two kinds of lectures:
 focus on technologies and theory (Pattern Recognition, Hidden
Markov Models, Automated Speech Recognition, Spoken Dialog
Systems, etc);
 focus on different aspects of VoiceXML
 For most of the semester we will alternate between the two kinds
of lectures on a weekly basis.
 The course material will be entirely self-contained
Requirements and Grading
 Fall 2006:
 There will be 5-8 assignments.
 Some of the assignments will be research to be presented
in class.
 Attendance is mandatory.
 Passing grade from the ethics class is required to pass
this course.
 Spring 2007:
 Group meeting
 Project evaluation
 Project report
Text
 McTear, Michael, Spoken Dialogue Technology Towards the Conversational User Interface. Springer
Verlag, 2004
Webpage
 http://www-cs.ccny.cuny.edu/~esther/capstone
 Visit before every lecture for the latest announcements
What is Voice XML?





VoiceXML
Architecture
Voice XML basics
Speech Recognition
Text-To-Speech
Download