Automatic Generation of Non-uniform HMM Topologies Based on the MDL Criterion

Takatoshi JITSUHIRO; Tomoko MATSUI; Satoshi NAKAMURA

Automatic Generation of Non-uniform HMM Topologies Based on the MDL Criterion

Takatoshi JITSUHIRO, Tomoko MATSUI, Satoshi NAKAMURA

Full Text Views

0

Cite this

Summary :

We propose a new method to introduce the Minimum Description Length (MDL) criterion to the automatic generation of non-uniform, context-dependent HMM topologies. Phonetic decision tree clustering is widely used, based on the Maximum Likelihood (ML) criterion, and only creates contextual variations. However, the ML criterion needs to predetermine control parameters, such as the total number of states, empirically for use as stop criteria. Information criteria have been applied to solve this problem for decision tree clustering. However, decision tree clustering cannot create topologies with various state lengths automatically. Therefore, we propose a method that applies the MDL criterion as split and stop criteria to the Successive State Splitting (SSS) algorithm as a means of generating contextual and temporal variations. This proposed method, the MDL-SSS algorithm, can automatically create adequate topologies without such predetermined parameters. Experimental results for travel arrangement dialogs and lecture speech show that the MDL-SSS can automatically stop splitting and obtain more appropriate HMM topologies than the original one.

Publication: IEICE TRANSACTIONS on Information Vol.E87-D No.8 pp.2121-2129

Publication Date: 2004/08/01

Publicized

Online ISSN

DOI

Type of Manuscript: PAPER

Category: Speech and Hearing

Cite this

Copy

Takatoshi JITSUHIRO, Tomoko MATSUI, Satoshi NAKAMURA, "Automatic Generation of Non-uniform HMM Topologies Based on the MDL Criterion" in IEICE TRANSACTIONS on Information, vol. E87-D, no. 8, pp. 2121-2129, August 2004, doi: .
Abstract: We propose a new method to introduce the Minimum Description Length (MDL) criterion to the automatic generation of non-uniform, context-dependent HMM topologies. Phonetic decision tree clustering is widely used, based on the Maximum Likelihood (ML) criterion, and only creates contextual variations. However, the ML criterion needs to predetermine control parameters, such as the total number of states, empirically for use as stop criteria. Information criteria have been applied to solve this problem for decision tree clustering. However, decision tree clustering cannot create topologies with various state lengths automatically. Therefore, we propose a method that applies the MDL criterion as split and stop criteria to the Successive State Splitting (SSS) algorithm as a means of generating contextual and temporal variations. This proposed method, the MDL-SSS algorithm, can automatically create adequate topologies without such predetermined parameters. Experimental results for travel arrangement dialogs and lecture speech show that the MDL-SSS can automatically stop splitting and obtain more appropriate HMM topologies than the original one.
URL: https://global.ieice.org/en_transactions/information/10.1587/e87-d_8_2121/_p

Copy

@ARTICLE{e87-d_8_2121,
author={Takatoshi JITSUHIRO, Tomoko MATSUI, Satoshi NAKAMURA, },
journal={IEICE TRANSACTIONS on Information},
title={Automatic Generation of Non-uniform HMM Topologies Based on the MDL Criterion},
year={2004},
volume={E87-D},
number={8},
pages={2121-2129},
abstract={We propose a new method to introduce the Minimum Description Length (MDL) criterion to the automatic generation of non-uniform, context-dependent HMM topologies. Phonetic decision tree clustering is widely used, based on the Maximum Likelihood (ML) criterion, and only creates contextual variations. However, the ML criterion needs to predetermine control parameters, such as the total number of states, empirically for use as stop criteria. Information criteria have been applied to solve this problem for decision tree clustering. However, decision tree clustering cannot create topologies with various state lengths automatically. Therefore, we propose a method that applies the MDL criterion as split and stop criteria to the Successive State Splitting (SSS) algorithm as a means of generating contextual and temporal variations. This proposed method, the MDL-SSS algorithm, can automatically create adequate topologies without such predetermined parameters. Experimental results for travel arrangement dialogs and lecture speech show that the MDL-SSS can automatically stop splitting and obtain more appropriate HMM topologies than the original one.},
keywords={},
doi={},
ISSN={},
month={August},}

Copy

TY - JOUR
TI - Automatic Generation of Non-uniform HMM Topologies Based on the MDL Criterion
T2 - IEICE TRANSACTIONS on Information
SP - 2121
EP - 2129
AU - Takatoshi JITSUHIRO
AU - Tomoko MATSUI
AU - Satoshi NAKAMURA
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E87-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2004
AB - We propose a new method to introduce the Minimum Description Length (MDL) criterion to the automatic generation of non-uniform, context-dependent HMM topologies. Phonetic decision tree clustering is widely used, based on the Maximum Likelihood (ML) criterion, and only creates contextual variations. However, the ML criterion needs to predetermine control parameters, such as the total number of states, empirically for use as stop criteria. Information criteria have been applied to solve this problem for decision tree clustering. However, decision tree clustering cannot create topologies with various state lengths automatically. Therefore, we propose a method that applies the MDL criterion as split and stop criteria to the Successive State Splitting (SSS) algorithm as a means of generating contextual and temporal variations. This proposed method, the MDL-SSS algorithm, can automatically create adequate topologies without such predetermined parameters. Experimental results for travel arrangement dialogs and lecture speech show that the MDL-SSS can automatically stop splitting and obtain more appropriate HMM topologies than the original one.
ER -