Electrical and Computer Engineering 797:
Speech and Audio Processing
(Winter 2004)

Objective:

To provide an introduction to basic concepts and methodologies for the analysis, modeling, synthesis and coding of speech and music. To provide a foundation for developing applications and for further study in the field.  To introduce software tools for the analysis and manipulation of speech and music and to gain practical experience in the design and implementation of speech and music processing algorithms.

Instructor:

Dr. Ian Bruce,
CRL 229, Ext. 26984.
ibruce@mail.ece.mcmaster.ca

Textbook:

• Quatieri, Discrete-Time Speech Signal Processing: Principles and Practice, 2001, Prentice Hall
  Copies available at Titles Bookstore.

References:

• Gold and Morgan, Speech and Audio Signal Processing: Processing and Perception of Speech and Music, 1999, John Wiley & Sons
• Deller, Hansen and Proakis, Discrete-Time Processing of Speech Signals, 1993, Macmillan (Reprinted in 2000 by Wiley-Interscience and IEEE Press)
• Oppenheim, Shafer and Buck, Discrete-time Signal Processing (2nd edition),1998, Prentice Hall.
• Roederer, The Physics and Psychophysics of Music: An Introduction, 1995, Springer-Verlag.
• Olson, Music, Physics and Engineering, 1967, Dover Publications.
• Gardner, The Virtual Acoustic Room, 1992, Master's Thesis, MIT Media Lab.

Course Outline:

Introduction to the Production and Classification of Speech Sounds (1.5 lectures)

• Introduction to the Speech Communication Pathway
• Anatomy and Physiology of Speech Production
• Spectrographic Analysis of Speech
• Categorization of Speech Sounds
• Prosody
• Speech Perception

Acoustics of Music Production (0.5 lectures)

• Physics of Sound
• Vibration of strings
• Resonance of tubes
• Pitch and Timbre

Acoustics of Speech Production (0.5 lectures)

• Uniform Tube Model
• Discrete-Time Modelling Based on Tube Concatenation
• Vocal Fold/Vocal Tract Interaction

Room Acoustics and Digital Effects (0.5 lectures)

• Sound Waves in Rooms
• Room Acoustics as a Component in Speech Systems
• Digital Effects

Analysis and Synthesis of Pole-Zero Speech Models (2 lectures)

• Time-Dependent Processing
• All-Pole Modeling of Deterministic Signals
• Linear Prediction Analysis of Stochastic Speech Sounds
• Synthesis Based on All-Pole Modeling

Short-Time Fourier Transform Analysis and Synthesis (1 lecture)

• Short-Time Analysis and Synthesis
• Signal Estimation from the Modified STFT
• Time-Scale Modification and Enhancement of Speech

Sinusoidal Analysis/Synthesis (0.5 lectures)

• Sinusoidal Speech Model
• Estimation of Sinewave Parameters
• Synthesis
• Source/Filter Phase Model
• Additive Deterministic-Stochastic Model

Homomorphic Signal Processing (2 lectures)

• Homomorphic Systems for Convolution
• Complex Cepstrum of Speech-Like Sequences
• Spectral Root Homomorphic Filtering
• Short-Time Homomorphic Analysis of Periodic Sequences
• Short-Time Speech Analysis
• Analysis/Synthesis Structures

Filter-Bank Analysis/Synthesis (1 lecture)

• Phase Vocoder
• Constant-Q Analysis/Synthesis
• Auditory Modeling

Speech Coding (1.5 lectures)

• Statistical Models of Speech
• Scalar Quantization
• Vector Quantization (VQ)
• Frequency-Domain Coding
• Model-Based Coding
• LPC Residual Coding

(Total Course = 11 x 3-hour lectures)

Term: II.

Assessment:

Assignments (30% ); Project and presentation (40%); Final (30%).

Lecture slides:

Lecture #1 (Tuesday, January 13)
Lecture #2 (Thursday, January 22) - Continues with notes from Lecture #1
Lecture #3 (Thursday, January 29)
Lecture #4 (Thursday, February 5)
Lecture #5 (Thursday, February 12)
Lecture #6 (Thursday, February 19) - Continues with notes from Lecture #5
Lecture #7 (Thursday, March 4)
Lecture #8 (Thursday, March 11)
Lecture #9 (Thursday, March 18) - Continues with notes from Lecture #8
Lecture #10 (Thursday, March 25)
Lecture #11 (Thursday, April 1)

ASSIGNMENTS:

Assignment #1 and required sound file oilyrag.wav.
Assignment #2 and required sound file assgn2.wav.
Assignment #3

Last updated Thursday, April 29, 2004