A MRF-Based Parallel Processing for Speech Recognition Using Linear Predictive HMM

Hideki NODA; Mehdi N. SHIRAZI; Mamoru NAKATSUI

A MRF-Based Parallel Processing for Speech Recognition Using Linear Predictive HMM

Hideki NODA, Mehdi N. SHIRAZI, Mamoru NAKATSUI

Full Text Views

0

Cite this

Summary :

Parallel processing in speech recognition is described, which is carried out at each frame on time axis. We have already proposed a parallel processing algorithm for HMM (Hidden Markov Model)-based speech recognition using Markov Random Fields (MRF). The parallel processing is realized by modeling the hidden state sequence by an MRF and using the Iterated Conditional Modes (ICM) algorithm to estimate the optimal state sequence given an observation sequence and model parameters. However this parallel processing with the ICM algorithm is applicable only to the standard HMM but not to the improved HMM like the linear predictive HMM which takes into account the correlations between nearby observation vectors. In this paper we propose a parallel processing algorithm applicable to the correlation-considered HMM, where a new deterministic relaxation algorithm called the Generalized ICM (GICM) algorithm is used instead of the ICM algorithm for estimation of the optimal state sequence. Speaker independent isolated word recognition experiments show the effectiveness of the proposed parallel processing using the GICM algorithm.

Publication: IEICE TRANSACTIONS on Information Vol.E77-D No.10 pp.1142-1147

Publication Date: 1994/10/25

Publicized

Online ISSN

DOI

Type of Manuscript: PAPER

Category: Speech Processing

Cite this

Copy

Hideki NODA, Mehdi N. SHIRAZI, Mamoru NAKATSUI, "A MRF-Based Parallel Processing for Speech Recognition Using Linear Predictive HMM" in IEICE TRANSACTIONS on Information, vol. E77-D, no. 10, pp. 1142-1147, October 1994, doi: .
Abstract: Parallel processing in speech recognition is described, which is carried out at each frame on time axis. We have already proposed a parallel processing algorithm for HMM (Hidden Markov Model)-based speech recognition using Markov Random Fields (MRF). The parallel processing is realized by modeling the hidden state sequence by an MRF and using the Iterated Conditional Modes (ICM) algorithm to estimate the optimal state sequence given an observation sequence and model parameters. However this parallel processing with the ICM algorithm is applicable only to the standard HMM but not to the improved HMM like the linear predictive HMM which takes into account the correlations between nearby observation vectors. In this paper we propose a parallel processing algorithm applicable to the correlation-considered HMM, where a new deterministic relaxation algorithm called the Generalized ICM (GICM) algorithm is used instead of the ICM algorithm for estimation of the optimal state sequence. Speaker independent isolated word recognition experiments show the effectiveness of the proposed parallel processing using the GICM algorithm.
URL: https://global.ieice.org/en_transactions/information/10.1587/e77-d_10_1142/_p

Copy

@ARTICLE{e77-d_10_1142,
author={Hideki NODA, Mehdi N. SHIRAZI, Mamoru NAKATSUI, },
journal={IEICE TRANSACTIONS on Information},
title={A MRF-Based Parallel Processing for Speech Recognition Using Linear Predictive HMM},
year={1994},
volume={E77-D},
number={10},
pages={1142-1147},
abstract={Parallel processing in speech recognition is described, which is carried out at each frame on time axis. We have already proposed a parallel processing algorithm for HMM (Hidden Markov Model)-based speech recognition using Markov Random Fields (MRF). The parallel processing is realized by modeling the hidden state sequence by an MRF and using the Iterated Conditional Modes (ICM) algorithm to estimate the optimal state sequence given an observation sequence and model parameters. However this parallel processing with the ICM algorithm is applicable only to the standard HMM but not to the improved HMM like the linear predictive HMM which takes into account the correlations between nearby observation vectors. In this paper we propose a parallel processing algorithm applicable to the correlation-considered HMM, where a new deterministic relaxation algorithm called the Generalized ICM (GICM) algorithm is used instead of the ICM algorithm for estimation of the optimal state sequence. Speaker independent isolated word recognition experiments show the effectiveness of the proposed parallel processing using the GICM algorithm.},
keywords={},
doi={},
ISSN={},
month={October},}

Copy

TY - JOUR
TI - A MRF-Based Parallel Processing for Speech Recognition Using Linear Predictive HMM
T2 - IEICE TRANSACTIONS on Information
SP - 1142
EP - 1147
AU - Hideki NODA
AU - Mehdi N. SHIRAZI
AU - Mamoru NAKATSUI
PY - 1994
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E77-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 1994
AB - Parallel processing in speech recognition is described, which is carried out at each frame on time axis. We have already proposed a parallel processing algorithm for HMM (Hidden Markov Model)-based speech recognition using Markov Random Fields (MRF). The parallel processing is realized by modeling the hidden state sequence by an MRF and using the Iterated Conditional Modes (ICM) algorithm to estimate the optimal state sequence given an observation sequence and model parameters. However this parallel processing with the ICM algorithm is applicable only to the standard HMM but not to the improved HMM like the linear predictive HMM which takes into account the correlations between nearby observation vectors. In this paper we propose a parallel processing algorithm applicable to the correlation-considered HMM, where a new deterministic relaxation algorithm called the Generalized ICM (GICM) algorithm is used instead of the ICM algorithm for estimation of the optimal state sequence. Speaker independent isolated word recognition experiments show the effectiveness of the proposed parallel processing using the GICM algorithm.
ER -