CSc 83300- Readings in Media Processing Fall 2006: SPEECH AND NATURAL LANGUAGE PROCESSING
Time:	Thursday 2-4 pm	Place:	3-308
Professor:	Prof. Esther Levin	Office Hours:
Email:	esther@cs.ccny.cuny.edu	Phone:	212 650-5626

Description:

The goal of this reading course is to introduce the students to a wide range of issues concerning the creation of computer programs that can interpret, generate, and learn natural language, both spoken and as text. Natural language processing has been considered one of the "holy grails" for artificial intelligence ever since Turing proposed his famed "imitation game" (the Turing Test). This course will focus on  mostly statistical approaches to language processing, and will illustrate the use of such methods in a variety of text- and speech-based application areas, including speech recognition, spoken dialogue systems, word-sense disambiguation,  machine translation, and text summarization.

Requirements and Grading:

The course will consist of lectures by the instructor (20-30%), talks by a few invited speakers from both academia and industry doing active research in Speech and Natural Language Processing (about 10 - 20%),  and reading presentations by students (60%).

For each student reading presentation class, a student presenter will be assigned to lead the discussion of a set of papers/book chapters on a given topic. The remaining students will email one or more questions per paper to the presenter, by 9 am in the morning of Wednesday before the class. These questions should be the kind of questions that you would ask if you heard the contents of the paper in a talk, or were reviewing the paper. The presenter will then email the collection of questions to everyone, which you should read (and print out) before class.

In addition to leading one or more class discussions, all students will be expected to do all the readings, and send the email questions as well as participate in the other discussions. Attendance is compulsory and absence will be penalized.

Grade Basis: email questions (20%), class participation (20%), leading 1 or 2 classes (60%),

Text:

We will be reading chapters from several textbooks and additional technical papers.

"Foundations of Statistical Natural Language Processing" by Manning & Schütze.

SPEECH and LANGUAGE PROCESSING: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, by D. Jurafsky and J.H. Martin,

Spoken Language Processing - A Guide to Theory, Algorithm, and System Development, by X. Huang, A. Acero, and H.W. Hon.

Spoken Dialog Technology: towards the conversational user interface, by Michael F. McTear

Fundamentals of Speech Recognition, by L.R. Rabiner and B.W. Juang,.

Topics:

1. Statistical Machine Translation( M&S- 13; J&M- 21)
2. Information Retrieval (M&S-15; J&M-17)
3. Text Categorization (M&S-16)
4. Collocations (M&S -5)
5. Word Sense Disambiguation (M&S-7)
6. Named Entity Recognition
7. Spoken Language Structure(HA&H -2)
8. Hidden Markov Models and Speech Recognition(M&S-9; J&M –7)
9. Discourse (J&M –18)
10. Dialogue and Conversational Agents (J&M-19)
11. Spoken Language Understanding (HA&H – 17)
12. Spoken Dialog Management (McTear – 13(4))

Announcements:

· By Monday September 4-th, please look at the list of topics, and send me an ordered list of topics that you would like to present, as well as any dates that you absolutely cannot lead class. If this constraint satisfaction process proves intractable, I will randomly assign students to topics/dates.

Guest Lectures:

September 7-th “Visions, technology, and business of conversational machines “, Dr. Roberto Pieraccini, SpeechCycle.

October 12-th: “Multimodality in Human-Computer Interfaces”, Dr. Michael Johnston, Dr. Srinivas Bangalore, AT&T Labs Research

November 16-th: “Automatic Text Summarization”, Prof. Noemie Elhadad , CCNY.

Schedule (Tentative and subject to change):

Date	Topic	Reading Material	Additional Material	Presenter
Aug 31-st	Preliminaries and Introduction	Mainly J&M – 1; Odyssey 2001 trailer ; Audio for the AT&T Communicator demo.		Prof. Levin
Sep. 7-th	Visions, technology, and business of conversational machines			Dr. Pieraccini
Sep 14-th	Spoken Language Structure	HA&H -2		Todd
Sep 21-st	Discourse	J&M-18	Example of Hobbs algorithm, A Statistical Approach to Anaphora Resolution	Sumon
Sep 28-th	Dialogue and Conversational Agents	J&M-22		Rachel
Oct 5-th	Hidden Markov Models and Speech Recognition	M&S-9; J&M –7	J&M-9 Hidden Markov Models, Tutorial on HMMs Larry Rabiner’s tutorial on HMM’s	Zheng
Oct 12-th	Multimodality in Human-Computer Interfaces			Dr. Michael Johnston, Dr. Srinivas Bangalore, AT&T Labs Research
Oct 19-th	Machine Transliteration	Joshua will present the following three papers, plus perhaps a fourth if there is time: These three cover three aspects of machine transliteration. The first, http://acl.ldc.upenn.edu/N/N06/N06-1060.pdf is about transliterating Arabic names for the purposes of information retreival of English documents containing transliterated Arabic names. The second, " Transliteration of proper names in cross-language applications" http://portal.acm.org/citation.cfm?id=860503&dl=ACM&coll=&CFID=15151515&CFTOKEN=6184618 is about forward transliteration from English names into Chinese. The third, "Machine Transliteration," http://coblitz.codeen.org:3125/citeseer.ist.psu.edu/cache/papers/cs/385/http:zSzzSzwww.isi.eduzSznatural-languagezSzmtzSztransliterate.pdf/knight97machine.pdf is about back-transliteration, and figuring out what the original English term was that was transliterated into Japanese, for the purpose of translating Japanese into English. (since many foreign terms and names have been transliterated into Japanese katakana) The fourth covers a statistical n-gram approach, which is mentioned in the first paper. http://portal.acm.org/citation.cfm?id=956890&dl=ACM&coll=&CFID=15151515&CFTOKEN=6184618		Joshua
Oct 26-th	Collocations	M&S-5		David
Nov 2-nd	Named Entity Recognition	Andrew Borthwick , A Maximum Entropy Approach to Named Entity Recognition (1999)		Tiziana
Nov 9-th	Information Retrieval	M&S-15; J&M-17		Lijun
Nov 16-th	Automatic Text Summarization			Prof. Noemie Elhadad , CCNY.
Nov 30-th	Machine Translation	M&S- 13; J&M- 24		Olga
Dec 7-th	Text Categorization	M&S-16		Minhua

Class Participants

Name	Email
Olga Lopusiewicz	olga.lopusiewicz*AT*gmail*DOT*com
Lijun Feng	Fenglj*AT*fastmail*DOT*fm
Joshua Waxman	Joshwaxman*AT*gmail*DOT*com
David Guy Brizan	Dbrizan*AT*gc*DOT*cuny*DOT*edu
Sumon Azhar	mqazhar*AT*sci*DOT*brooklyn*DOT*cuny*DOT*edu
Rachel Adler	rachelfadler*AT*gmail*DOT*com
Zheng Chen	chenfuqing*AT*gmail*DOT*com
Tiziana Ligorio	TLigorio*AT*gc*DOT*cuny*DOT*edu
Todd Flyr	TFlyr*AT*bmcc*DOT*cuny*DOT*edu
Minhua Huang	minhuahuang2003*AT*yahoo.com
Fay Halberstam	fhalberstam*AT*gc*DOT*cuny*DOT*edu

Useful or Interesting Links:

ACL SIGdial

Special Issue on Intelligent Dialogue Systems for the ETAI area Intelligent User Interfaces