Speech Processing by Computer







By the end of the session you should:

be able to describe the basic architecture of an isolated word recognition system

be able to list the main sources of variability which present problems for such systems and how they are typically overcome

be able to describe what end-point detection does

be able to justify what acoustic parameters are used for matching

be able to explain why non-linear time alignment is necessary and how it works in general terms




9.      Isolated Word Recognition

9.1.   Aims

9.1.1.      Word identification

9.2.   Applications

9.2.1.      Command and Control

9.3.   Problems

9.3.1.      Segmentation words

9.3.2.      Speaker variability dependency

9.3.3.      Environmental variability microphone

9.3.4.      Discrimination vocabulary

9.4.   Construction

9.4.1.      End-point detection

9.4.2.      Acoustic Processing shape parameters

9.4.3.      Matching distance alignment



G. Chollet, "Automatic Speech and Speaker Recognition", in Fundamentals of Speech Synthesis and Speech Recognition ed E. Keller, Wiley, 1994.


J.N. Holmes, "Speech recognition by pattern matching of whole words", in Speech Synthesis and Recognition, van Nostrand, 1988.