Affective Spoken Dialogue Systems

(CS 3710 / ISSP 3565: Advanced Topics in Artificial Intelligence)
Time: Tu Th 11:00-12:15  Place 6516 Sennott Square
Professor:  Diane Litman Office Hours: Tu 10-11 (Th by appt), Tu 1:30-3:30, Th 2:30-3 (5105 Sennott Square)
Email:  litman at cs dot pitt dot edu Phone:  412-624-8838 (Sennott Square); 412-624-1261 (LRDC)

Description:

Human conversational partners are able, at least to a certain extent, to detect the speaker's or listener's emotional state and may attempt to respond to it accordingly. When instead one of the interlocutors is a computer a number of questions arise, such as the following: To what extent are dialogue systems able to simulate such behaviors? Can we learn the mechanisms of emotional behaviors from observing and analyzing the behavior of human speakers? How can emotions be automatically recognized from a user's mimics, gestures and speech? What possibilities does a dialogue system have to express emotions itself? And, very importantly, would emotional system behavior be desirable at all? [Preface to Affective Dialogue Systems, 2004]
This course will explore topics in the area of Affective Spoken Dialogue Systems. The course will both introduce foundational areas such as Affective Computing, Emotional Speech, and Spoken Dialogue Systems, and present recent work addressing their integration. While the course will primarily be presented from a Speech and Natural Language Processing perspective, papers from related areas will also be discussed. Topics to be covered will include:

Prerequisites:

Graduate-level courses in Natural Language Processing or Artificial Intelligence, or permission of instructor.

Syllabus (subject to change!):

Schedule

Readings

Course Requirements:

Course Project: 40% (proposal: 8%, presentation: 8%, report: 24%)

Class Presentations: 25%

Reaction Essays : 25%

Class Participation: 10%

Because in-class discussions are an important part of this course, absences are strongly discouraged. Even if an absence is unavoidable, you are still responsible for making arrangements to turn in the assignments on time. Late assignments will not be accepted. No extensions will be given, except in case of extraordinary and documented circumstances (e.g., hospitalization).

Academic Integrity:

If you include material from any source in your presentation and/or projects, you must acknowledge it. Your presentations and projects should represent your (and your partner's, if applicable) original work.

Students with Disabilities:

If you have a disability for which you are or may be requesting an accommodation, you are encouraged to contact both your instructor and Disability Resources and Services, 216 William Pitt Union, (412) 648-7890/(412) 383-7355 (TTY), as early as possible in the term. DRS will verify your disability and determine reasonable accomodations for this course.