Detail výsledku
Feature And Score Level Combination Of Subspace Gaussians In LVCSR Task
We have demonstrated that the SGMM framework is an efficient approachin the LVCSR task. Overall evaluations of SGMMs exploitingpowerful but complex PLP-BN features yield similar results asthose obtained by conventional HMM/GMMs. Nevertheless, the totalnumber of SGMM parameters is about 3 times less than in theHMM/GMM framework. Evaluation results also indicate differentproperties of the examined acoustic modeling techniques. AlthoughSGMMs consistently outperform HMM/GMMs when built over individualfeatures, HMM/GMMs can benefit much more from thefeature-level combination than SGMMs. Nevertheless based on ananalysis measuring complementarity of individual recognition systems,we show that SGMM-based recognizers produce heterogeneousoutputs (scores) and thus subsequent score-level combinationcan bring additional improvement.
Automatic Speech Recognition, Discriminativefeatures, System combination
@inproceedings{BUT103519,
author="Petr {Motlíček} and Daniel {Povey} and Martin {Karafiát}",
title="Feature And Score Level Combination Of Subspace Gaussians In LVCSR Task",
booktitle="Proceedings of ICASSP 2013",
year="2013",
pages="7604--7608",
publisher="IEEE Signal Processing Society",
address="Vancouver",
isbn="978-1-4799-0355-9",
url="http://www.fit.vutbr.cz/research/groups/speech/publi/2013/motlicek_icassp2013_0007604.pdf"
}