Julia Hirschberg

CS 4995/6998: Topics in Speech Processing: Computational Approaches to Emotional Speech, Fall 2011

Time: W 6:10pm-8:00pm
Place: CSB 453 (CS Conference Room)

Professor Julia Hirschberg (Office Hours: Tu 3:30-4:30, Th 2:30-3:30)
julia@cs.columbia.edu, 212-939-7114

Teaching Assistant TBD

Description

This course introduces students to research on emotional speech and its applications in call centers, tutoring systems, and medical diagnosis. We will explore state-of-the-art work on recognizing and producing classic emotions automatically. Emotions such as anger, happiness, sadness, and uncertainty are important to recognize in online dialogue systems and to produce in computer games. We will also study the recognition of other types of speaker state, including deceptive and charismatic speech, and the uses of acoustic and prosodic information in medical domains, for the diagnosis of mental and physical conditions. Classes will be lecture and discussion with an emphasis on group participation. There are no prerequisites for the course and no exams.

Students will be expected to complete all required reading assignments before the class for which they are assigned. They will prepare weekly discussion questions based on assigned readings and submit them by noon on the day before class. Prepare 3 questions for each of the required papers for the week. Each student will lead one class discussion on an optional paper of their choice. They will also prepare a term Project either alone or in teams for presentation during the finals period assigned for the course. There will be no midterm or final exam.

NB: The Columbia Speech Lab is available for use in projects on a signup basis.

Grading

Class participation: 30%

Class presentation: 10%

Project Proposal: 10%

Interim Report: 10%

Full project submission: 30%

Project presentation: 10%

Academic Integrity

Copying or paraphrasing someone's work (code included), or permitting your own work to be copied or paraphrased, even if only in part, is not allowed, and will result in an automatic grade of 0 for the entire assignment or exam in which the copying or paraphrasing was done. Your grade should reflect your own work. If you believe you are going to have trouble completing an assignment, please talk to the instructor or TA in advance of the due date.

Date	Topic	Readings	Due Dates
Week 1 (9/7)	Introduction	Emotional Synthetic Speech; The Love Detector
Week 2 (9/14)	Evidence of Emotion: Theory, Experiments and Corpora	Cornelius00; Schereretal 03 Wiltingetal06; Enos&Hirschberg06
Week 3 (9/21)	Cues to Emotional Language; Introduction to Acoustic/Prosodic Features;Introduction to Praat	Liscombeetal05; *Kumaretal06; Praat tutorial; Acoustics of Speech; Speech sounds	Download Praat to your laptop and bring headphones if you have them
Week 4 (9/28)	Techniques for Emotion Classification	Cowie00; Feeltrace; Whissel's Dictionary of Affect; Satohetal07; Nisimuraelal06;	Project Ideas due
Week 5 (10/5)	Emotion in IVR Systems: Anger and Frustration	Devillers&Vidrascu06; Angetal02; Laukkaetal11 (Wilkey); Gupta&Rajput07
Week 6 (10/12)	Emotion in Tutoring Systems: Confidence and Confusion	Aietal06 (Shin); Liscombeetal05b	Project Proposal due
Week 7 (10/19)	Emotion in Meetings: Hot Spots and Laughter	Wrede&Shriberg03; Laskowski&Burger07 (Graham); *Jurafskyetal09 (Willson)
Week 8 (10/26)	Producing Emotional Speech	Cahn90; Eideetal04; Beskow&Nordenberg06; Yanushevskayaetal05 (Kuo); cf Synface demo, Mary TTS, JHU workshops
Week 9 (11/2)	Cultural Differences	Braun&Katerbow05; Ekman&Friesen03; *Shahidetal07 (Park);	Interim Report due
Week 10 (11/9)	Lexical Cues to Emotion: Sentiment Analysis	Hatzivassiloglouetal97(Satsangi); Pangetal02;Turney02; Cohnetal04 (Ritko); Agarwaletal11(Kripa); DAL
Week 11 (11/16)	Non-Linguistic Cues to Emotion: Face, Body, Brain	Johnstoneetal06,Ekman video; Ekman bio; cf FACS
Week 12 (Mon, 11/21)	Deception Detection	Hirschbergetal05; Eriksson&Lacerda07 (Cooper); YouDecide;Hancocketal08 (Cabral); DePaulo03
Week 13 (11/30)	Emotion Detection in Medical Diagnosis	Hirschbergetal10; Levitetal01; Ramirez-Esparzaetal08; Kalioubyetal06; Rudeetal04 (Ulinski); Snowdenetal; Luoetal06
Week 14 (12/7)	Modeling Other Speaker States: Sarcasm, Charisma; "Good news, bad news"	Teppermanetal06; Davidovetal10; Biadsyetal08; Paralinguistic Challenge 2009, *2010, 2011; Swerts&Hirschberg10	Projects due
Finals week (12/16-12/23)			Project presentations