Detail publikace
Some like it Gaussian ...
MATĚJKA, P., SCHWARZ, P., KARAFIÁT, M., ČERNOCKÝ, J.
Originální název
Some like it Gaussian ...
Anglický název
Some like it Gaussian ...
Jazyk
en
Originální abstrakt
In Hidden Markov models, speech features are modeled by Gaussian distributions. In this paper, we propose to gaussianize the features to better fit to this modeling. A distribution of the data is estimated and a transform function is derived. We have tested two methods of the transform estimation (global and speaker based). The results are reported on recognition of isolated Czech words (SpeechDat-E) with CI and CD models and on medium vocabulary continuous speech recognition task (SPINE). Gaussianized data provided in all three cases results superior to standard MFC coefficients proving, that the gaussianization is a cheap way to increase the recognition accuracy.
Anglický abstrakt
In Hidden Markov models, speech features are modeled by Gaussian distributions. In this paper, we propose to gaussianize the features to better fit to this modeling. A distribution of the data is estimated and a transform function is derived. We have tested two methods of the transform estimation (global and speaker based). The results are reported on recognition of isolated Czech words (SpeechDat-E) with CI and CD models and on medium vocabulary continuous speech recognition task (SPINE). Gaussianized data provided in all three cases results superior to standard MFC coefficients proving, that the gaussianization is a cheap way to increase the recognition accuracy.
Dokumenty
BibTex
@inproceedings{BUT4282,
author="Pavel {Matějka} and Petr {Schwarz} and Martin {Karafiát} and Jan {Černocký}",
title="Some like it Gaussian ...",
annote="In Hidden Markov models, speech features are modeled by Gaussian distributions. In this paper, we propose to gaussianize the features to better fit to this modeling. A distribution of the data is estimated and a transform function is derived. We have tested two methods of the transform estimation (global and speaker based). The results are reported on recognition of isolated Czech words (SpeechDat-E) with CI and CD models and on medium vocabulary continuous speech recognition task (SPINE). Gaussianized data provided in all three cases results superior to standard MFC coefficients proving, that the gaussianization is a cheap way to increase the recognition accuracy.",
booktitle="Proceedings of the conference TSD'2002",
chapter="4282",
year="2002",
month="september",
pages="321",
type="conference paper"
}