Pronounced ``Why nought''. It was one of the decoders used in the WERNICKE project and is still in use to make phonetic alignments.

In IR work this matrix is weighted by terms designed to enhance retrieval; there seems no good reason for weighting the unigram probability estimates in language modelling work.

It is indeed known that transitions are more important to speech perception than stationary parts.

thanks to a suggestion of John Bridle.

Jean-Marc Boite
Tue Jan 7 12:46:31 MET 1997