Paper Info
Title
Improvements on the pronunciation prefix tree search organization.
Abstract
The need for ever more efficient search organizations persists as the size and complexity of the knowledge sources used in continuous speech recognition (CSR) tasks continues to increase. We address efficiency issues associated with a search organization based on pronunciation prefix trees (PPTs). In particular we present (1) a mechanism that eliminates redundant computations in non-reentrant trees, (2) a comparison of two methods for distributing language model probabilities in PPTs, and (3) report results on two look ahead pruning strategies. Using the 1994 DARPA 20 k NAB word bigram for the male segment of si dev5m 92 (the 5k speaker independent development test set for the WSJ), the error rate was 12.2% with a real-time factor of 1.0 on a 120 MHz Pentium.
Year
DOI
Venue
1996
10.1109/ICASSP.1996.540308
ICASSP
Keywords
Field
DocType
pronunciation prefix tree search,search organization,male segment,mhz pentium,error rate,efficient search organization,efficiency issue,knowledge source,language model probability,continuous speech recognition,k nab word bigram,lattices,real time factor,probability,efficiency,decoding,pentium processor,history,speech recognition,natural languages,look ahead,distributed computing,real time,system testing,language model
Pronunciation,Computer science,Word error rate,Speech recognition,Prefix,Bigram,Pentium,Artificial intelligence,Natural language processing,Trie,Language model,Test set
Conference
ISBN
Citations 
PageRank 
0-7803-3192-3
19
4.42
References 
Authors
5
3
Name
Order
Citations
PageRank
F. Alleva132780.97
Xuedong Huang21390283.19
Mei-Yuh Hwang3477124.33