Concordance

Query H90-1060 33 >
Random sample 33 (1,056.5 per million)

tech,7-1-H90-1060,ak	This paper reports on two contributions to <term> large vocabulary continuous speech recognition </term> .	#21042 This paper reports on two contributions tolarge vocabulary continuous speech recognition.
tech,8-2-H90-1060,ak	First , we present a new paradigm for <term> speaker-independent ( SI ) training </term> of <term> hidden Markov models ( HMM ) </term> , which uses a large amount of <term> speech </term> from a few <term> speakers </term> instead of the traditional practice of using a little speech from many <term> speakers </term> .	#21056 First, we present a new paradigm forspeaker-independent ( SI ) training of hidden Markov models (HMM), which uses a large amount of speech from a few speakers instead of the traditional practice of using a little speech from many speakers.
model,14-2-H90-1060,ak	First , we present a new paradigm for <term> speaker-independent ( SI ) training </term> of <term> hidden Markov models ( HMM ) </term> , which uses a large amount of <term> speech </term> from a few <term> speakers </term> instead of the traditional practice of using a little speech from many <term> speakers </term> .	#21062 First, we present a new paradigm for speaker-independent (SI) training ofhidden Markov models ( HMM ), which uses a large amount of speech from a few speakers instead of the traditional practice of using a little speech from many speakers.
lr,27-2-H90-1060,ak	First , we present a new paradigm for <term> speaker-independent ( SI ) training </term> of <term> hidden Markov models ( HMM ) </term> , which uses a large amount of <term> speech </term> from a few <term> speakers </term> instead of the traditional practice of using a little speech from many <term> speakers </term> .	#21075 First, we present a new paradigm for speaker-independent (SI) training of hidden Markov models (HMM), which uses a large amount ofspeech from a few speakers instead of the traditional practice of using a little speech from many speakers.
other,31-2-H90-1060,ak	First , we present a new paradigm for <term> speaker-independent ( SI ) training </term> of <term> hidden Markov models ( HMM ) </term> , which uses a large amount of <term> speech </term> from a few <term> speakers </term> instead of the traditional practice of using a little speech from many <term> speakers </term> .	#21079 First, we present a new paradigm for speaker-independent (SI) training of hidden Markov models (HMM), which uses a large amount of speech from a fewspeakers instead of the traditional practice of using a little speech from many speakers.
other,44-2-H90-1060,ak	First , we present a new paradigm for <term> speaker-independent ( SI ) training </term> of <term> hidden Markov models ( HMM ) </term> , which uses a large amount of <term> speech </term> from a few <term> speakers </term> instead of the traditional practice of using a little speech from many <term> speakers </term> .	#21092 First, we present a new paradigm for speaker-independent (SI) training of hidden Markov models (HMM), which uses a large amount of speech from a few speakers instead of the traditional practice of using a little speech from manyspeakers.
other,6-3-H90-1060,ak	In addition , combination of the <term> training speakers </term> is done by averaging the statistics of independently trained <term> models </term> rather than the usual <term> pooling </term> of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> .	#21100 In addition, combination of thetraining speakers is done by averaging the statistics of independently trained models rather than the usual pooling of all the speech data from many speakers prior to training.
model,17-3-H90-1060,ak	In addition , combination of the <term> training speakers </term> is done by averaging the statistics of independently trained <term> models </term> rather than the usual <term> pooling </term> of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> .	#21111 In addition, combination of the training speakers is done by averaging the statistics of independently trainedmodels rather than the usual pooling of all the speech data from many speakers prior to training.
tech,22-3-H90-1060,ak	In addition , combination of the <term> training speakers </term> is done by averaging the statistics of independently trained <term> models </term> rather than the usual <term> pooling </term> of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> .	#21116 In addition, combination of the training speakers is done by averaging the statistics of independently trained models rather than the usualpooling of all the speech data from many speakers prior to training.
lr,26-3-H90-1060,ak	In addition , combination of the <term> training speakers </term> is done by averaging the statistics of independently trained <term> models </term> rather than the usual <term> pooling </term> of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> .	#21120 In addition, combination of the training speakers is done by averaging the statistics of independently trained models rather than the usual pooling of all thespeech data from many speakers prior to training.
other,30-3-H90-1060,ak	In addition , combination of the <term> training speakers </term> is done by averaging the statistics of independently trained <term> models </term> rather than the usual <term> pooling </term> of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> .	#21124 In addition, combination of the training speakers is done by averaging the statistics of independently trained models rather than the usual pooling of all the speech data from manyspeakers prior to training.
tech,33-3-H90-1060,ak	In addition , combination of the <term> training speakers </term> is done by averaging the statistics of independently trained <term> models </term> rather than the usual <term> pooling </term> of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> .	#21127 In addition, combination of the training speakers is done by averaging the statistics of independently trained models rather than the usual pooling of all the speech data from many speakers prior totraining.
other,3-4-H90-1060,ak	With only 12 <term> training speakers </term> for <term> SI recognition </term> , we achieved a 7.5 % <term> word error rate </term> on a standard <term> grammar </term> and <term> test set </term> from the <term> DARPA Resource Management corpus </term> .	#21132 With only 12training speakers for SI recognition, we achieved a 7.5% word error rate on a standard grammar and test set from the DARPA Resource Management corpus.
tech,6-4-H90-1060,ak	With only 12 <term> training speakers </term> for <term> SI recognition </term> , we achieved a 7.5 % <term> word error rate </term> on a standard <term> grammar </term> and <term> test set </term> from the <term> DARPA Resource Management corpus </term> .	#21135 With only 12 training speakers forSI recognition, we achieved a 7.5% word error rate on a standard grammar and test set from the DARPA Resource Management corpus.
measure(ment),14-4-H90-1060,ak	With only 12 <term> training speakers </term> for <term> SI recognition </term> , we achieved a 7.5 % <term> word error rate </term> on a standard <term> grammar </term> and <term> test set </term> from the <term> DARPA Resource Management corpus </term> .	#21143 With only 12 training speakers for SI recognition, we achieved a 7.5%word error rate on a standard grammar and test set from the DARPA Resource Management corpus.
lr,20-4-H90-1060,ak	With only 12 <term> training speakers </term> for <term> SI recognition </term> , we achieved a 7.5 % <term> word error rate </term> on a standard <term> grammar </term> and <term> test set </term> from the <term> DARPA Resource Management corpus </term> .	#21149 With only 12 training speakers for SI recognition, we achieved a 7.5% word error rate on a standardgrammar and test set from the DARPA Resource Management corpus.
other,22-4-H90-1060,ak	With only 12 <term> training speakers </term> for <term> SI recognition </term> , we achieved a 7.5 % <term> word error rate </term> on a standard <term> grammar </term> and <term> test set </term> from the <term> DARPA Resource Management corpus </term> .	#21151 With only 12 training speakers for SI recognition, we achieved a 7.5% word error rate on a standard grammar andtest set from the DARPA Resource Management corpus.
lr-prod,26-4-H90-1060,ak	With only 12 <term> training speakers </term> for <term> SI recognition </term> , we achieved a 7.5 % <term> word error rate </term> on a standard <term> grammar </term> and <term> test set </term> from the <term> DARPA Resource Management corpus </term> .	#21155 With only 12 training speakers for SI recognition, we achieved a 7.5% word error rate on a standard grammar and test set from theDARPA Resource Management corpus.
other,10-5-H90-1060,ak	This performance is comparable to our best condition for this <term> test suite </term> , using 109 <term> training speakers </term> .	#21170 This performance is comparable to our best condition for thistest suite, using 109 training speakers.
other,15-5-H90-1060,ak	This performance is comparable to our best condition for this <term> test suite </term> , using 109 <term> training speakers </term> .	#21175 This performance is comparable to our best condition for this test suite, using 109training speakers.


	in Help