SYLLABUS
Contact Information:

Time	MF: 2:15 - 3:30 PM
Place	432 Simrall
Overseer	Joseph Picone Office: 413 Simrall Office Hours: 11-12 MWF (others by appt.) Email: picone@cavs.msstate.edu
Class Alias	ece_8000@cavs.msstate.edu
URL	http://www.cavs.msstate.edu/research/isip/publications/courses/ece_8000_speech

Grading Policies:

Presentation No. 1	24.99%
Presentation No. 2	25.01%
Paper No. 1	25.00%
Paper No. 2	25.00%

Presentations:

All presentations will be HTML-based following the formats provided. A typical presentation will include: the motivation for the topic - relevance to speech recognition; a detailed explanation of the underlying theory and fundamental concepts; examples demonstrating these concepts; and a short section explaining applications of this technology to speech recognition and citing any experimental evidence of its effectiveness. The presentation grade will be based on the effectiveness of the presentation, and the degree to which technical questions were answered in a convincing manner.

Papers:

Each student will prepare an 8 page companion paper for each presentation. This paper will be prepared in a standard IEEE ICASSP conference-style format, and follow all format specifications expected of such documents (use ICASSP example as a template).

Schedule:

Class	Date	Topic	Lecturer
1	10/25	Principle of Maximum Entropy	Aravind Ganapathiraju
2	10/29	LDA, PCA, and ICA	Vishwanath Mantha
3	11/01	Fundamentals of Linear Prediction	Shivali Srivastava
4	11/05	Dynamic Programming	Ramasubramanian Sundaram
5	11/08	???Factor Analysis???	Vishwanath Mantha
6	11/12	The Impact of Cross-Entropy and Mutual Information on Human Language Engineering	Jie Zhao
7	11/15	Learning Bayesian Networks: The Combination of Knowledge and Statistical Data	Yufeng Wu
8	11/19	Iterative Algorithms	Ramasubramanian Sundaram
9	11/22	Linear Regression	Shivali Srivastava
10	11/29	Speaker Adaptation Techniques for LVCSR	Jon Hamaker
11	12/03	Discriminative Techniques in HMM Training	Aravind Ganapathiraju
12	12/06	Introduction to Finite State Machines	Jie Zhao
13	12/09	Belief Networks	Yufeng Wu
14	12/10	Markov Random Fields	Jon Hamaker
15	12/16	All	Final Papers Due