Syllabus: Topics in Information Systems, Spring 2015
Southern Oregon University
Department of Computer Science
CS
415 (CRN 7071) / CS 515 (CRN 7081), Four Credits
Prerequisite: CS 258
Instructor Information
Instructor: Dan Harvey
Room: Computer Science Building #CS218
Phone: 552-6149
E-mail: harveyd@sou.edu
Office Hours: Mon/Wed: 12:30-1:30, Tue/Thu 10:30-12:30
Web
Site: http://cs.sou.edu/~harveyd
The web site
is available for quiz results, lab assignments, weekly handouts, current grade
status, and contact with class members. Click on the appropriate class, and
then select the desired option.
Class Times: Tuesday, Thursday: 8:30-10:20 EDPSY 164
Final Exam: Thursday, June 11,
8:00-10:00
Course Text: Fundamentals
of Speech Recognition, Rabiner and Juang, Prentice Hall, ISBN 978-0-13-015157-5
Catalog Description: Explores selected topics in information systems. Sample topics include database systems, networking and the Internet, and creating business frameworks.
Course Objectives: We will focus on the concepts necessary for building a framework for automatic speech recognition (ASR). This is an emerging application area, which among many uses, is critical for providing user friendly interfaces to mobile devices. We will review digital signal processing concepts and then explore how these relate to extracting audio features necessary for building a front end for ASR. Topics related to back end processing include Neural Networks, Hidden Markov Models (HMM), Dynamic Time Warping, Dynamic Programming, Transducers and Finite State Machines, N-gram models, and others.
Course Objectives
Students will achieve a solid introduction to the design issues and popular algorithms necessary for building an ASR framework, but this knowledge also has many uses beyond this particular application area. Sphinx, an open source ASR project, and the ACORNS language acquisition software package, whose purpose is to revitalize indigenous languages, will be platforms that we will utilize as to enhance student learning. These also expose students to non-trivial real-world implementations.
1. Introduction to the discipline of natural language processing. Digital signal processing background
2. Feature extraction, digital filters
3. Cepstral and Linear Prediction analysis
4. Noise suppression, voice activity detection
5. Clustering techniques
6. Dynamic time warping (DTW), feature normalization
7. Hidden Markov Models (HMM)
8. N Grams, smoothing techniques
9. Language models and morphology
10. Language
models and grammars, part of speech tagging
Course Grading
There will be three exams where the low score is dropped for 40% and a final for an additional 30% of the total grade. There will be three labs and a final presentation totaling 30% of the total grade. The last lab will involve a student presentation.
Grade Breakdown: 93-100% A 90-92% A-
88-89% B+ 82-87 B 80-81% B-
78-79% C+ 72-77 C 70-71% C-
68-69% D+ 62-67 D 60-61% D-
Under 60 F
Disabilities:
If you are
in need of academic support because of a documented disability (whether it be
learning, mobility, psychiatric, health-related, or sensory) you may be
eligible for academic or other accommodations through Disability Services for
Students. Contact Disability Services for Students; Director DSS 552-6213, or
schedule an appointment in person at the ACCESS Center, Stevenson Union,
lower-level.
Culture
of supporting and reporting:
SOU has a wide range of resources to help you
succeed. Our faculty, staff, and administration are dedicated to providing you
with the best possible support. The SOU Cares Report allows us to connect you
with staff members who can assist with concerns, including financial, health,
mental health, wellbeing, legal concerns, family concerns, harassment, assault,
study skills, time management, etc. You are also welcome to use the SOU Cares
Report to share concerns about yourself, a friend, or a classmate at
www.sou.edu/ssi. These concerns can include reports related to academic
integrity, harassment, bias, or assault. Reports related to sexual misconduct
or sexual assault can be made anonymously or confidentially. Student Support
and Intervention provides recourse for students through the Student Code of
Conduct, Title IX, Affirmative Action, and other applicable policies,
regulations, and laws.
Mandatory
reporting:
Federal law requires that employees of institutions
of higher learning (faculty, staff and administrators) report to a Title IX officer any time they
become aware that a student is a victim or perpetrator of gender-based bias,
sexual harassment, sexual assault, domestic violence, or stalking. Further,
Oregon law requires a mandatory report to law enforcement of any physical or
emotional abuse of a child or other protected person, including elders and
people with disabilities. If you are the victim of sexual or physical abuse and
wish to make a confidential disclosure please contact any of SOU’s confidential
advisors (http://www.sou.edu/ssi/confidential-advisors.html), or use Southern
Oregon University's Anonymous Harassment, Violence, and Interpersonal
Misconduct Reporting Form https://jfe.qualtrics.com/form/SV_7R7CCBciGNL473L