A hybrid neural system for phonematic transformation

Igor T. Podolak, Seong Whan Lee, Andrzej Bielecki, Elzbieta Majkut

Research output: Chapter in Book/Report/Conference proceedingChapter

5 Citations (Scopus)

Abstract

Text-to-phoneme conversion is a common problem in speech processing. This can be done using a rule-based system or a neural network. In this paper we propose a solution to this problem using a modular hybrid system that uses basic rules to subdivide the original problem into easier tasks which are then solved by dedicated neural networks. Such a solution can be more rapidly constructed, and is easily extendable. A voting committee concept is used to enhance generalization abilities of the system.

Original languageEnglish
Title of host publicationProceedings - International Conference on Pattern Recognition
Pages957-960
Number of pages4
Volume15
Edition2
Publication statusPublished - 2000

Fingerprint

Neural networks
Speech processing
Knowledge based systems
Hybrid systems

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture

Cite this

Podolak, I. T., Lee, S. W., Bielecki, A., & Majkut, E. (2000). A hybrid neural system for phonematic transformation. In Proceedings - International Conference on Pattern Recognition (2 ed., Vol. 15, pp. 957-960)

A hybrid neural system for phonematic transformation. / Podolak, Igor T.; Lee, Seong Whan; Bielecki, Andrzej; Majkut, Elzbieta.

Proceedings - International Conference on Pattern Recognition. Vol. 15 2. ed. 2000. p. 957-960.

Research output: Chapter in Book/Report/Conference proceedingChapter

Podolak, IT, Lee, SW, Bielecki, A & Majkut, E 2000, A hybrid neural system for phonematic transformation. in Proceedings - International Conference on Pattern Recognition. 2 edn, vol. 15, pp. 957-960.
Podolak IT, Lee SW, Bielecki A, Majkut E. A hybrid neural system for phonematic transformation. In Proceedings - International Conference on Pattern Recognition. 2 ed. Vol. 15. 2000. p. 957-960
Podolak, Igor T. ; Lee, Seong Whan ; Bielecki, Andrzej ; Majkut, Elzbieta. / A hybrid neural system for phonematic transformation. Proceedings - International Conference on Pattern Recognition. Vol. 15 2. ed. 2000. pp. 957-960
@inbook{cbd639ebcecd46c2b7781b832c1ef8b0,
title = "A hybrid neural system for phonematic transformation",
abstract = "Text-to-phoneme conversion is a common problem in speech processing. This can be done using a rule-based system or a neural network. In this paper we propose a solution to this problem using a modular hybrid system that uses basic rules to subdivide the original problem into easier tasks which are then solved by dedicated neural networks. Such a solution can be more rapidly constructed, and is easily extendable. A voting committee concept is used to enhance generalization abilities of the system.",
author = "Podolak, {Igor T.} and Lee, {Seong Whan} and Andrzej Bielecki and Elzbieta Majkut",
year = "2000",
language = "English",
volume = "15",
pages = "957--960",
booktitle = "Proceedings - International Conference on Pattern Recognition",
edition = "2",

}

TY - CHAP

T1 - A hybrid neural system for phonematic transformation

AU - Podolak, Igor T.

AU - Lee, Seong Whan

AU - Bielecki, Andrzej

AU - Majkut, Elzbieta

PY - 2000

Y1 - 2000

N2 - Text-to-phoneme conversion is a common problem in speech processing. This can be done using a rule-based system or a neural network. In this paper we propose a solution to this problem using a modular hybrid system that uses basic rules to subdivide the original problem into easier tasks which are then solved by dedicated neural networks. Such a solution can be more rapidly constructed, and is easily extendable. A voting committee concept is used to enhance generalization abilities of the system.

AB - Text-to-phoneme conversion is a common problem in speech processing. This can be done using a rule-based system or a neural network. In this paper we propose a solution to this problem using a modular hybrid system that uses basic rules to subdivide the original problem into easier tasks which are then solved by dedicated neural networks. Such a solution can be more rapidly constructed, and is easily extendable. A voting committee concept is used to enhance generalization abilities of the system.

UR - http://www.scopus.com/inward/record.url?scp=10244222152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=10244222152&partnerID=8YFLogxK

M3 - Chapter

VL - 15

SP - 957

EP - 960

BT - Proceedings - International Conference on Pattern Recognition

ER -