EECS 225d (3 units)
Audio Signal Processing in Humans and Machines
Initial class (Jan. 22): Cory Hall 400 (Hughes Room), Wed 4:00-5:30
Afterwards: ICSI, 1947 Center Street, 6th floor, MW 4:00-5:30
Spring Semester, 2014
Professor Morgan (with contributions from several area experts)
The focus of this recently modified course is on engineering models for speech and audio processing, with a new emphasis on student participation in presentations. Mirroring the more recent edition of the associated textbook, the class will include material about current audio processing methods such as psychoacoustic audio coding (e.g., MP3) and sound source separation. The methods discussed are used to design systems for analysis, synthesis, and recognition of speech and music. For many of these topics we will discuss not only the engineering methods, but also some of the physiological and psychoacoustic properties of the human auditory and speech generation systems. This latter information can provide an important perspective: how can we make use of knowledge about these natural systems when we design artificial ones?
Topics will include, among others: auditory physiology; room acoustics; music signal analysis; speech synthesis; models of speech production and perception; signal processing for speech analysis; robustness to environmental variation in speech recognizers; statistical speech recognition, including introduction to Hidden Markov Model and discriminative approaches.
Previous classes have included students from both EE and CS, whose varied backgrounds have enriched the course. In other words, even though it's an EE course, be not afraid, CS students!
As noted above. student participation will be a key part of this year's course. In particular, after the introductory lectures, students will prepare presentations that build on the class slide packs, which will be made available prior to the class, and we will focus on discussion on the relevant topics. These classes will be augmented by talks by visiting experts on particular subtopics. The course concludes with presentations by the students describing their course projects.
Prerequisites: EE123 or equivalent, and Stat 200A or equivalent; or grad standing and consent of instructor
Grading: 60% for final project, 20% for homeworks and quizzes in the first half, and 20% for class participation via student presentations of material in the middle of the course.
Required text: B. Gold, N. Morgan, and D. Ellis ``Speech and Audio Signal Processing'', 2nd edition, Wiley Press 2011.
Supplementary reading:
EECS 225d Tentative Schedule, Spring 2014
HISTORICAL BACKGROUND
WEEK 1:
1. Overall Introduction: What's the big idea? Jan 22 (Morgan)
WEEK 2:
2. Brief History of Synthetic Audio/speech analysis and synthesis Jan 27 (Morgan)
3. Brief History of Automatic Speech Recognition Jan 29 (Morgan)
WEEK 3:
4. Speech Recognition Overview Feb 3 (Morgan)
5. Speech Recognition Continued Feb 5 (Morgan)
SCIENTIFIC BACKGROUND
WEEK 4:
6. Human Speech Recognition Feb 10 (Morgan)
7. Pattern Classification Feb 12 (Morgan)
WEEK 5:
Holiday, Feb 17
8. Statistical Pattern Classification Feb 19 (Morgan)
WEEK 6:
9. Acoustical Basics Feb 24 (Students)
10. Linguistic sound categories Feb 26 (Johnson, Linguistics UCB)
WEEK 7:
11. Neural Networks for speech March 3 (Students)
12. Room Acoustics March 5 (Morgan)
ENGINEERING APPLICATIONS
WEEK 8:
13. Feature Extraction for ASR 1 March 10 (Morgan)
14. Feature Extraction for ASR 2 March 12 (Students)
WEEK 9:
15. Deterministic Sequence recognition for ASR March 17 (Students)
16. Statistical Sequence recognition and training for ASR March 19 (Students)
WEEK 10:
Spring Break March 24-28
WEEK 11:
17. Adaptation and Discriminant Training for ASR Mar 31 (Students)
18. Perceptual Audio Coding April 2 (Lazzaro)
WEEK 12:
19. Class Cancelled April 7 (Students)
20. Speech Synthesis April 9 (Silverman, Apple)
WEEK 13:
21. Source Separation April 14 (Students)
22. Music Signal Analysis April 16 (Students)
WEEK 14:
23. Auditory physiology 1 April 21 (Ghitza)
24. Auditory physiology 2 April 23 (Ghitza)
WEEK 15:
25. Hearing 1 April 28 (Ghitza)
26. Hearing 2 April 30 (Ghitza)
WEEK 16:
27. Speaker Verification May 5 (Students)
28. Student presentations (extended session, recitation week) May 7