On the other side , we use nearest neighbor approximation to calculate gussian mixture densities , which can reduce recognition time by 6 . 67 % compared with standard viterbi beam search algorithm 另一方面,使用高斯混合概率密度的最近鄰快速估算方法,使標(biāo)準(zhǔn)viterbibeam搜索算法的搜索速度提高了6 . 67 。
Further more , we improve the nearest neighbor approximation method by calculat e mixtures ordered by likelihood of being the best scoring mixture . the likelihood is calculating from previously processed data . this improved method can reduce recognition time by 15 . 56 % compared with standard viterbi beam search algorithm 本文對(duì)最近鄰快速估算方法進(jìn)行改進(jìn),在搜索過(guò)程中根據(jù)已處理過(guò)的數(shù)據(jù)統(tǒng)計(jì)出各個(gè)高斯混合分量產(chǎn)生最高對(duì)數(shù)概率的概率,并依此預(yù)測(cè)隨后的計(jì)算中最有可能產(chǎn)生最高對(duì)數(shù)概率的高斯混合分量,優(yōu)先計(jì)算更有可能產(chǎn)生最高對(duì)數(shù)概率的高斯混合分量,使標(biāo)準(zhǔn)viterbibeam搜索算法的搜索速度提高了15 . 56 。