Sixth European Conference on Speech Communication and Technology

Budapest, Hungary
September 5-9, 1999

Speed Improvement of the Time-Asynchronous Acoustic Fast Match

Miroslav Novak, Michael Picheny

IBM T.J. Watson Research Center Yorktown Heights, NY, USA

This paper describes an algorithm for improvement of the speed of a time-asynchronous fast match, which is a part of a stack-search based recognition system. This fast match uses a phonetic tree to represent the entire vocabulary of the recognizer. Evaluation of the tree (in a depth-first manner), can be done much more eficiently using the fact that under certain conditions, the results of branch evaluations can be used to approximate the scores of other branches of the tree.

Full Paper (PDF)   Gnu-Zipped Postscript

Bibliographic reference.  Novak, Miroslav / Picheny, Michael (1999): "Speed improvement of the time-asynchronous acoustic fast match", In EUROSPEECH'99, 1115-1118.