In this paper a signal modeling technique based upon finite mixture autoregressive probabilistic functions of Markov chains is developed and applied to the problem of speech recognition, particularly speaker-independent recognition of isolated digits. Two types of mixture probability densities are investigated: finite mixtures of Gaussian autoregressive densities (GAM) and nearest-neighbor partitioned finite mixtures of Gaussian autoregressive densities (PGAM). In the former (GAM), the observation density in each Markov state is simply a (stochastically constrained) weighted sum of Gaussian autoregressive densities, while in the latter (PGAM) it involves nearest-neighbor decoding which in effect, defines a set of partitions on the observation space. In this paper we discuss the signal modeling methodology and give experimental results on speaker independent recognition of isolated digits. We also discuss the potential use of the modeling technique for other applications.
%0 Journal Article
%1 Juang1985
%A Juang, Biing-Hwang
%A Rabiner, Lawrence R.
%D 1985
%J IEEE Transactions on Acoustics, Speech and Signal Processing
%K Decoding;Density Markov analysis;Spectral analysis;Speech distribution;Signal estimation;Parameter estimation;Probability functional likelihood models;Maximum processes recognition;Stochastic theory;Hidden
%N 6
%P 1404-1413
%R 10.1109/TASSP.1985.1164727
%T Mixture autoregressive hidden Markov models for speech signals
%V 33
%X In this paper a signal modeling technique based upon finite mixture autoregressive probabilistic functions of Markov chains is developed and applied to the problem of speech recognition, particularly speaker-independent recognition of isolated digits. Two types of mixture probability densities are investigated: finite mixtures of Gaussian autoregressive densities (GAM) and nearest-neighbor partitioned finite mixtures of Gaussian autoregressive densities (PGAM). In the former (GAM), the observation density in each Markov state is simply a (stochastically constrained) weighted sum of Gaussian autoregressive densities, while in the latter (PGAM) it involves nearest-neighbor decoding which in effect, defines a set of partitions on the observation space. In this paper we discuss the signal modeling methodology and give experimental results on speaker independent recognition of isolated digits. We also discuss the potential use of the modeling technique for other applications.
@article{Juang1985,
abstract = {In this paper a signal modeling technique based upon finite mixture autoregressive probabilistic functions of Markov chains is developed and applied to the problem of speech recognition, particularly speaker-independent recognition of isolated digits. Two types of mixture probability densities are investigated: finite mixtures of Gaussian autoregressive densities (GAM) and nearest-neighbor partitioned finite mixtures of Gaussian autoregressive densities (PGAM). In the former (GAM), the observation density in each Markov state is simply a (stochastically constrained) weighted sum of Gaussian autoregressive densities, while in the latter (PGAM) it involves nearest-neighbor decoding which in effect, defines a set of partitions on the observation space. In this paper we discuss the signal modeling methodology and give experimental results on speaker independent recognition of isolated digits. We also discuss the potential use of the modeling technique for other applications.},
added-at = {2021-02-01T10:51:23.000+0100},
author = {Juang, Biing-Hwang and Rabiner, Lawrence R.},
biburl = {https://www.bibsonomy.org/bibtex/2822ef70ba27b3385c1dea03efe31e9ab/m-toman},
doi = {10.1109/TASSP.1985.1164727},
file = {:pdfs/juang_transassp_1985.pdf:PDF},
interhash = {e3841f1f282bdb82d56060104f62b6f5},
intrahash = {822ef70ba27b3385c1dea03efe31e9ab},
issn = {0096-3518},
journal = {IEEE Transactions on Acoustics, Speech and Signal Processing},
keywords = {Decoding;Density Markov analysis;Spectral analysis;Speech distribution;Signal estimation;Parameter estimation;Probability functional likelihood models;Maximum processes recognition;Stochastic theory;Hidden},
month = dec,
number = 6,
owner = {schabus},
pages = {1404-1413},
timestamp = {2021-02-01T10:51:23.000+0100},
title = {Mixture autoregressive hidden Markov models for speech signals},
volume = 33,
year = 1985
}