LECTURE 22: FUNDAMENTALS OF MARKOV MODELS

• Objectives:
  
  
  • Introduce a Markov model
    
    
  • Understand the difference between an observable and a hidden Markov model
    
    
  • Appreciate the reason we use Markov models: to model temporal evolution of the spectrum (important in speech recognition!)
    
    
  • Demonstrate basic calculations
    
    
  • Demonstrate the infeasibility of these basic calculations for real problems

This material can be found in most speech recognition and pattern recognition textbooks. These notes follow material presented in:

J. Deller, et. al., Discrete-Time Processing of Speech Signals, MacMillan Publishing Co., ISBN: 0-7803-5386-2, 2000.

Another useful reference is:

L.R. Rabiner and B.W. Juang, Fundamentals of Speech Recognition, Prentice-Hall, ISBN: 0-13-015157-2, 1993.

The course textbook also follows these traditional references closely.