Title :
The time-conditioned approach in dynamic programming search for LVCSR
Author :
Ortmanns, Stefan ; Ney, Hermann
Author_Institution :
Dept. of Comput. Sci., Tech. Hochschule Aachen, Germany
fDate :
11/1/2000 12:00:00 AM
Abstract :
This paper presents the time-conditioned approach in dynamic programming search for large-vocabulary continuous-speech recognition. The following topics are presented: the baseline algorithm, a time-synchronous beam search version, a comparison with the word-conditioned approach, a comparison with stack decoding. The approach has been successfully tested on the NAB task using a vocabulary of 64000 words
Keywords :
decoding; dynamic programming; search problems; speech coding; speech recognition; synchronisation; LVCSR; NAB task; baseline algorithm; dynamic programming search; large-vocabulary continuous-speech recognition; one-pass DP search; search organization; stack decoding; time-conditioned approach; time-synchronous beam search version; word-conditioned approach; Acoustic emission; Computer science; Decoding; Dynamic programming; Hidden Markov models; History; Search problems; Speech recognition; Testing; Vocabulary;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
