W10-1717 will certainly experiment with lattice pruning in future . Powell 's algorithm
H89-1012 the lattice size through various lattice pruning tech - niques . We have used
W03-1727 parsing is three times slower when lattice pruning is turned off . • The parser
W03-1727 scores are later used in word lattice pruning and parse ranking ( Wu and Jiang
H89-1012 computation tractable , we applied the lattice pruning techniques described above to
P00-1018 and looking at various levels of lattice pruning and ways to incorporate recogniser
J05-3002 alignment , overcutting during lattice pruning caused the omission of three
N09-1046 starting and ending vertices . 3.2 Lattice pruning Except for the minimum segment
W03-1727 the OOV rate increases . • Lattice pruning makes a big difference as well
W03-1727 to be done during parsing , the lattice pruning is non-greedy so that no plausible
H94-1064 bigram point of view ) and a word lattice pruning threshold used to control the
W03-1727 in Wu et al ( 2002 ) . 1.4 Word lattice pruning Now that all the possible words
J92-3001 acceptance lexicon for use in lattice pruning during speech processing . The
D11-1044 is shown as Alg . 1 . Pass 1 : Lattice Pruning After generating phrase lattices
W03-1727 new word identification , word lattice pruning and parsing . The final segmentation
W03-1727 ization " , " pruning " for " lattice pruning " and " tun - ing " for " tuning
J14-2005 place of the actual vine 6.3.2 Lattice Pruning . To reduce phrase lattice sizes
hide detail