LNGN450-01
Speech and Language Processing
Tuesday, 5:30-8
Dr. Joan Bachenko
j.bachenko@verizon.net

 

TEXT: Daniel Jurafsky and James H. Martin. Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition.

COURSE REQUIREMENTS:

You are responsible for all the material in the assigned readings, homework, class lectures and discussions.

The course assignments and their grade weight are as follows: five homework assignments worth 15% of the final grade, four midterm assignments worth 40%, and a final assignment worth 45%. Homework assignments will be graded as Pass/Fail. You will be given a Pass if you turn the assignment in.

You must present your final project to the class. The last class of the semester will be reserved for student presentations.

ASSIGNMENT SCHEDULE:

First Midterm due February 11th: normalize a text; run the stemmer, summarize results

Second Midterm due Feb 25th: participate in a team project on POS tagging

Third Midterm due March 18th: write a one-page proposal for final project

Fourth Midterm due April 1st: build a language model for speech recognition

COURSE SCHEDULE:

Jan. 14

Chapter 1, Review of course, introduction and overview of topics

Jan. 21

Chapter 2, begin Chapter 3
Regular expressions, morphological analysis, Unix tools
Homework #1: regular expressions, FSA’s

Jan. 28

Chapter 3
Morphological analysis, corpora, more Unix tools
Homework #2: tr, sort, uniq, grep

Feb. 4

Chapter 8
Words and part of speech

Feb. 11

Chapter 9, begin Chapter 10
Syntax: grammars and parsing
Homework #3: tag a small corpus, compare to POS tagger program
First Midterm is due.

Feb. 18

Chapter 10, begin Chapter 6
Syntactic parsing, ngram grammars

Feb. 25

Chapter 6
ngrams, language models for speech recognition
Second Midterm is due
Homework #4: language model statistics, perplexity

Mar. 4

Chapter 7
speech recognition, phonetics, language and acoustic models

Mar. 18

Chapter 7
speech recognition
Third Midterm is due.
Homework #5: try out a speech recognition system

Mar. 25

Chapter 4
Text-to-speech synthesis, rules, dictionaries, prosody

Apr. 1

Chapter 4, begin Chapter 14
TTS, meaning
Fourth Midterm is due.

Apr. 8

Chapters 14, 15
semantic analysis

Apr. 15

Applications and discussion

Apr. 22

Student presentations