Model parameter estimation for mixture density polynomial segment models

Toshiaki Fukada, Yoshinori Sagisaka, Kuldip K. Paliwal

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

20 Citations (Scopus)

Abstract

In this paper, we propose parameter estimation techniques for mixture density polynomial segment models (henceforth MDPSM) where their trajectories are specified with an arbitrary regression order. MDPSM parameters can be trained in one of three different ways: (1) segment clustering, (2) expectation maximization (EM) training of mean trajectories, or (3) EM training of mean and variance trajectories. These parameter estimation methods were evaluated in TIMIT vowel classification experiments. The experimental results showed that modeling both the mean and variance trajectories are consistently superior to modeling only the mean trajectory. We also found that modeling both trajectories results in significant improvements over the conventional HMM.

Original languageEnglish
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Editors Anon
PublisherIEEE
Pages1403-1406
Number of pages4
Volume2
Publication statusPublished - 1997
Externally publishedYes
EventProceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 5) - Munich, Ger
Duration: 1997 Apr 211997 Apr 24

Other

OtherProceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 5)
CityMunich, Ger
Period97/4/2197/4/24

Fingerprint

Parameter estimation
polynomials
Trajectories
trajectories
education
vowels
Statistical Models
regression analysis
Experiments

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Fukada, T., Sagisaka, Y., & Paliwal, K. K. (1997). Model parameter estimation for mixture density polynomial segment models. In Anon (Ed.), ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 2, pp. 1403-1406). IEEE.

Model parameter estimation for mixture density polynomial segment models. / Fukada, Toshiaki; Sagisaka, Yoshinori; Paliwal, Kuldip K.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. ed. / Anon. Vol. 2 IEEE, 1997. p. 1403-1406.

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

Fukada, T, Sagisaka, Y & Paliwal, KK 1997, Model parameter estimation for mixture density polynomial segment models. in Anon (ed.), ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 2, IEEE, pp. 1403-1406, Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 5), Munich, Ger, 97/4/21.
Fukada T, Sagisaka Y, Paliwal KK. Model parameter estimation for mixture density polynomial segment models. In Anon, editor, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2. IEEE. 1997. p. 1403-1406
Fukada, Toshiaki ; Sagisaka, Yoshinori ; Paliwal, Kuldip K. / Model parameter estimation for mixture density polynomial segment models. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. editor / Anon. Vol. 2 IEEE, 1997. pp. 1403-1406
@inproceedings{511ac51be03d4008a3c5c934293fe7ba,
title = "Model parameter estimation for mixture density polynomial segment models",
abstract = "In this paper, we propose parameter estimation techniques for mixture density polynomial segment models (henceforth MDPSM) where their trajectories are specified with an arbitrary regression order. MDPSM parameters can be trained in one of three different ways: (1) segment clustering, (2) expectation maximization (EM) training of mean trajectories, or (3) EM training of mean and variance trajectories. These parameter estimation methods were evaluated in TIMIT vowel classification experiments. The experimental results showed that modeling both the mean and variance trajectories are consistently superior to modeling only the mean trajectory. We also found that modeling both trajectories results in significant improvements over the conventional HMM.",
author = "Toshiaki Fukada and Yoshinori Sagisaka and Paliwal, {Kuldip K.}",
year = "1997",
language = "English",
volume = "2",
pages = "1403--1406",
editor = "Anon",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Model parameter estimation for mixture density polynomial segment models

AU - Fukada, Toshiaki

AU - Sagisaka, Yoshinori

AU - Paliwal, Kuldip K.

PY - 1997

Y1 - 1997

N2 - In this paper, we propose parameter estimation techniques for mixture density polynomial segment models (henceforth MDPSM) where their trajectories are specified with an arbitrary regression order. MDPSM parameters can be trained in one of three different ways: (1) segment clustering, (2) expectation maximization (EM) training of mean trajectories, or (3) EM training of mean and variance trajectories. These parameter estimation methods were evaluated in TIMIT vowel classification experiments. The experimental results showed that modeling both the mean and variance trajectories are consistently superior to modeling only the mean trajectory. We also found that modeling both trajectories results in significant improvements over the conventional HMM.

AB - In this paper, we propose parameter estimation techniques for mixture density polynomial segment models (henceforth MDPSM) where their trajectories are specified with an arbitrary regression order. MDPSM parameters can be trained in one of three different ways: (1) segment clustering, (2) expectation maximization (EM) training of mean trajectories, or (3) EM training of mean and variance trajectories. These parameter estimation methods were evaluated in TIMIT vowel classification experiments. The experimental results showed that modeling both the mean and variance trajectories are consistently superior to modeling only the mean trajectory. We also found that modeling both trajectories results in significant improvements over the conventional HMM.

UR - http://www.scopus.com/inward/record.url?scp=0030681725&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030681725&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0030681725

VL - 2

SP - 1403

EP - 1406

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

A2 - Anon, null

PB - IEEE

ER -