Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers

Youngkyu Cho, Sung A. Kim, Dongsuk Yook

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Today's state-of-the-art speech recognition systems typically use continuous density hidden Markov models with mixture of Gaussian distributions. Such speech recognition systems have problems; they require too much memory to run, and are too slow for large vocabulary applications. Two approaches are proposed for the design of compact acoustic models, namely, subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models. However, these models require also large memory to acquire high recognition accuracy. In this paper, we propose a new hybrid model using subspace distribution clustering hidden Markov model and semi-continuous hidden Markov model with the aim of achieving much more compact acoustic models.

Original languageEnglish
Title of host publication8th International Conference on Spoken Language Processing, ICSLP 2004
PublisherInternational Speech Communication Association
Pages669-672
Number of pages4
Publication statusPublished - 2004
Event8th International Conference on Spoken Language Processing, ICSLP 2004 - Jeju, Jeju Island, Korea, Republic of
Duration: 2004 Oct 42004 Oct 8

Other

Other8th International Conference on Spoken Language Processing, ICSLP 2004
CountryKorea, Republic of
CityJeju, Jeju Island
Period04/10/404/10/8

Fingerprint

acoustics
Hidden Markov Model
Hybrid Model
vocabulary
Speech Recognition
Acoustics
Vocabulary

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Cho, Y., Kim, S. A., & Yook, D. (2004). Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers. In 8th International Conference on Spoken Language Processing, ICSLP 2004 (pp. 669-672). International Speech Communication Association.

Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers. / Cho, Youngkyu; Kim, Sung A.; Yook, Dongsuk.

8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association, 2004. p. 669-672.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cho, Y, Kim, SA & Yook, D 2004, Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers. in 8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association, pp. 669-672, 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of, 04/10/4.
Cho Y, Kim SA, Yook D. Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers. In 8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association. 2004. p. 669-672
Cho, Youngkyu ; Kim, Sung A. ; Yook, Dongsuk. / Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers. 8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association, 2004. pp. 669-672
@inproceedings{20bc041ae5044987af98fe8c04e6c788,
title = "Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers",
abstract = "Today's state-of-the-art speech recognition systems typically use continuous density hidden Markov models with mixture of Gaussian distributions. Such speech recognition systems have problems; they require too much memory to run, and are too slow for large vocabulary applications. Two approaches are proposed for the design of compact acoustic models, namely, subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models. However, these models require also large memory to acquire high recognition accuracy. In this paper, we propose a new hybrid model using subspace distribution clustering hidden Markov model and semi-continuous hidden Markov model with the aim of achieving much more compact acoustic models.",
author = "Youngkyu Cho and Kim, {Sung A.} and Dongsuk Yook",
year = "2004",
language = "English",
pages = "669--672",
booktitle = "8th International Conference on Spoken Language Processing, ICSLP 2004",
publisher = "International Speech Communication Association",

}

TY - GEN

T1 - Hybrid model using subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models for embedded speech recognizers

AU - Cho, Youngkyu

AU - Kim, Sung A.

AU - Yook, Dongsuk

PY - 2004

Y1 - 2004

N2 - Today's state-of-the-art speech recognition systems typically use continuous density hidden Markov models with mixture of Gaussian distributions. Such speech recognition systems have problems; they require too much memory to run, and are too slow for large vocabulary applications. Two approaches are proposed for the design of compact acoustic models, namely, subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models. However, these models require also large memory to acquire high recognition accuracy. In this paper, we propose a new hybrid model using subspace distribution clustering hidden Markov model and semi-continuous hidden Markov model with the aim of achieving much more compact acoustic models.

AB - Today's state-of-the-art speech recognition systems typically use continuous density hidden Markov models with mixture of Gaussian distributions. Such speech recognition systems have problems; they require too much memory to run, and are too slow for large vocabulary applications. Two approaches are proposed for the design of compact acoustic models, namely, subspace distribution clustering hidden Markov models and semi-continuous hidden Markov models. However, these models require also large memory to acquire high recognition accuracy. In this paper, we propose a new hybrid model using subspace distribution clustering hidden Markov model and semi-continuous hidden Markov model with the aim of achieving much more compact acoustic models.

UR - http://www.scopus.com/inward/record.url?scp=85009141777&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009141777&partnerID=8YFLogxK

M3 - Conference contribution

SP - 669

EP - 672

BT - 8th International Conference on Spoken Language Processing, ICSLP 2004

PB - International Speech Communication Association

ER -