LECTURE 33: SMOOTHING N-GRAM LANGUAGE MODELS

• Objectives:
  
  
  • Why do we need N-gram smoothing?
    
    
  • Deleted interpolation
    
    
  • Backoff language models
    
    
  • Discounting

This lecture combines material from the course textbook:

X. Huang, A. Acero, and H.W. Hon, Spoken Language Processing - A Guide to Theory, Algorithm, and System Development, Prentice Hall, Upper Saddle River, New Jersey, USA, ISBN: 0-13-022616-5, 2001.

and from this source:

F. Jelinek, Statistical Methods for Speech Recognition, MIT Press, Boston, Massachusetts, USA, ISBN: 0-262-10066-5, 1998.