Automatic generation of multiple pronunciations based on neural networks

Toshiaki Fukada, Takayoshi Yoshimura, Yoshinori Sagisaka

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

We propose a method for automatically generating a pronunciation dictionary based on a pronunciation neural network that can predict plausible pronunciations (realized pronunciations) from the canonical pronunciation. This method can generate multiple forms of realized pronunciations using the pronunciation network. For generating a sophisticated realized pronunciation dictionary, two techniques are described: (1) realized pronunciations with likelihoods and (2) realized pronunciations for word boundary phonemes. Experimental results on spontaneous speech show that the automatically derived pronunciation dictionaries give consistently higher recognition rates than a conventional dictionary.

Original languageEnglish
Pages (from-to)63-73
Number of pages11
JournalSpeech Communication
Volume27
Issue number1
Publication statusPublished - 1999
Externally publishedYes

Fingerprint

Glossaries
neural network
dictionary
Neural Networks
Neural networks
Likelihood
Predict
Dictionary
Experimental Results

Keywords

  • Neural networks
  • Pronunciation dictionary
  • Speech recognition
  • Spontaneous speech

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Experimental and Cognitive Psychology
  • Linguistics and Language

Cite this

Automatic generation of multiple pronunciations based on neural networks. / Fukada, Toshiaki; Yoshimura, Takayoshi; Sagisaka, Yoshinori.

In: Speech Communication, Vol. 27, No. 1, 1999, p. 63-73.

Research output: Contribution to journalArticle

Fukada, Toshiaki ; Yoshimura, Takayoshi ; Sagisaka, Yoshinori. / Automatic generation of multiple pronunciations based on neural networks. In: Speech Communication. 1999 ; Vol. 27, No. 1. pp. 63-73.
@article{1dbce36e5bd74db4a3a4d181c241cd10,
title = "Automatic generation of multiple pronunciations based on neural networks",
abstract = "We propose a method for automatically generating a pronunciation dictionary based on a pronunciation neural network that can predict plausible pronunciations (realized pronunciations) from the canonical pronunciation. This method can generate multiple forms of realized pronunciations using the pronunciation network. For generating a sophisticated realized pronunciation dictionary, two techniques are described: (1) realized pronunciations with likelihoods and (2) realized pronunciations for word boundary phonemes. Experimental results on spontaneous speech show that the automatically derived pronunciation dictionaries give consistently higher recognition rates than a conventional dictionary.",
keywords = "Neural networks, Pronunciation dictionary, Speech recognition, Spontaneous speech",
author = "Toshiaki Fukada and Takayoshi Yoshimura and Yoshinori Sagisaka",
year = "1999",
language = "English",
volume = "27",
pages = "63--73",
journal = "Speech Communication",
issn = "0167-6393",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Automatic generation of multiple pronunciations based on neural networks

AU - Fukada, Toshiaki

AU - Yoshimura, Takayoshi

AU - Sagisaka, Yoshinori

PY - 1999

Y1 - 1999

N2 - We propose a method for automatically generating a pronunciation dictionary based on a pronunciation neural network that can predict plausible pronunciations (realized pronunciations) from the canonical pronunciation. This method can generate multiple forms of realized pronunciations using the pronunciation network. For generating a sophisticated realized pronunciation dictionary, two techniques are described: (1) realized pronunciations with likelihoods and (2) realized pronunciations for word boundary phonemes. Experimental results on spontaneous speech show that the automatically derived pronunciation dictionaries give consistently higher recognition rates than a conventional dictionary.

AB - We propose a method for automatically generating a pronunciation dictionary based on a pronunciation neural network that can predict plausible pronunciations (realized pronunciations) from the canonical pronunciation. This method can generate multiple forms of realized pronunciations using the pronunciation network. For generating a sophisticated realized pronunciation dictionary, two techniques are described: (1) realized pronunciations with likelihoods and (2) realized pronunciations for word boundary phonemes. Experimental results on spontaneous speech show that the automatically derived pronunciation dictionaries give consistently higher recognition rates than a conventional dictionary.

KW - Neural networks

KW - Pronunciation dictionary

KW - Speech recognition

KW - Spontaneous speech

UR - http://www.scopus.com/inward/record.url?scp=0033077780&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033077780&partnerID=8YFLogxK

M3 - Article

VL - 27

SP - 63

EP - 73

JO - Speech Communication

JF - Speech Communication

SN - 0167-6393

IS - 1

ER -