Computer Science 248/448

Speech Recognition and Statistical Natural Language Processing

Fall 2005

Instructor: Dan Gildea
TA: Hao Zhang
Location: TTh 4:50-6:05pm, CSB 632.

Homeworks

Text: Manning and Schuetze, Foundations of Statistical Natural Language Processing
Recommended: Jurafsky and Martin, Speech and Language Processing, Handouts from Prof. Allen

Syllabus

Onwe will coverwhich means that after class you will understandif before class you have read
9/1 Introduction empiricism MS ch 1
9/6 Probability Theory independence, information theory MS ch 2
9/8 Linguistic Essentials X-bar MS ch 3
9/13 N-grams deleted interpolation MS ch 6
9/15 N-grams Good Turing
9/20 Hidden Markov Models viterbi, forward-backward MS ch 9
9/22 Part of Speech Tagging EM, transformation-based learning MS ch 10
9/27 Signal Processing for Speech fourier, MFCC handout
9/29 Approaches to Speech Recognition putting it all together handout
10/4 Large Vocabulary Speech Recognition beam search, rescoring Jelinek ch 4, 5
10/6 Large Vocabulary Speech Recognition multistack decoding, A* Jelinek ch 4, 5
10/11 Collocations t-tests MS ch 5
10/13 Word Sense Disambiguation bootstrapping MS ch 7
10/18 Language Formalisms Swiss German sarkar
10/20 Parsing charts, inside-outside MS ch 11
10/25 Review
10/27 Midterm
11/1 Midterm Solutions
11/3 Real-World Parsing lexicalization MS ch 12
11/8 Machine Translation source-channel knight
11/10 Machine Translation model 1, 2, 3
11/15 Machine Translation decoding germann et al.
11/17 Machine Translation syntax-based MT yamada & knight
11/22 Machine Translation syntax-based decoding, inversion transduction grammar
11/29 Clustering k-means MS ch 14
12/1 Information Retrieval Latent Semantic Analysis MS ch 15
12/6 Text Categorization maximum entropy MS ch 16
12/8 Review come to class with questions!
Final Exam: Tuesday, December 20, 12:30pm

Grading


gildea @ cs rochester edu
December 6, 2005