Phoneme Set Design for Speech Recognition of English by Japanese

Xiaoyun WANG; Jinsong ZHANG; Masafumi NISHIDA; Seiichi YAMAMOTO

doi:10.1587/transinf.2014EDP7168

Phoneme Set Design for Speech Recognition of English by Japanese

Xiaoyun WANG, Jinsong ZHANG, Masafumi NISHIDA, Seiichi YAMAMOTO

Full Text Views

0

Share
Cite this

Summary :

This paper describes a novel method to improve the performance of second language speech recognition when the mother tongue of users is known. Considering that second language speech usually includes less fluent pronunciation and more frequent pronunciation mistakes, the authors propose using a reduced phoneme set generated by a phonetic decision tree (PDT)-based top-down sequential splitting method instead of the canonical one of the second language. The authors verify the efficacy of the proposed method using second language speech collected with a translation game type dialogue-based English CALL system. Experiments show that a speech recognizer achieved higher recognition accuracy with the reduced phoneme set than with the canonical phoneme set.

Publication: IEICE TRANSACTIONS on Information Vol.E98-D No.1 pp.148-156

Publication Date: 2015/01/01

Publicized: 2014/10/01

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2014EDP7168

Type of Manuscript: PAPER

Category: Speech and Hearing

Authors

Xiaoyun WANG
  Doshisha University
Jinsong ZHANG
  Beijing Language and Culture University
Masafumi NISHIDA
  Doshisha University
Seiichi YAMAMOTO
  Doshisha University

Keyword

phonetic decision tree (PDT), Phoneme set, Second language speech recognition

Cite this

Copy

Xiaoyun WANG, Jinsong ZHANG, Masafumi NISHIDA, Seiichi YAMAMOTO, "Phoneme Set Design for Speech Recognition of English by Japanese" in IEICE TRANSACTIONS on Information, vol. E98-D, no. 1, pp. 148-156, January 2015, doi: 10.1587/transinf.2014EDP7168.
Abstract: This paper describes a novel method to improve the performance of second language speech recognition when the mother tongue of users is known. Considering that second language speech usually includes less fluent pronunciation and more frequent pronunciation mistakes, the authors propose using a reduced phoneme set generated by a phonetic decision tree (PDT)-based top-down sequential splitting method instead of the canonical one of the second language. The authors verify the efficacy of the proposed method using second language speech collected with a translation game type dialogue-based English CALL system. Experiments show that a speech recognizer achieved higher recognition accuracy with the reduced phoneme set than with the canonical phoneme set.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2014EDP7168/_p

Copy

@ARTICLE{e98-d_1_148,
author={Xiaoyun WANG, Jinsong ZHANG, Masafumi NISHIDA, Seiichi YAMAMOTO, },
journal={IEICE TRANSACTIONS on Information},
title={Phoneme Set Design for Speech Recognition of English by Japanese},
year={2015},
volume={E98-D},
number={1},
pages={148-156},
abstract={This paper describes a novel method to improve the performance of second language speech recognition when the mother tongue of users is known. Considering that second language speech usually includes less fluent pronunciation and more frequent pronunciation mistakes, the authors propose using a reduced phoneme set generated by a phonetic decision tree (PDT)-based top-down sequential splitting method instead of the canonical one of the second language. The authors verify the efficacy of the proposed method using second language speech collected with a translation game type dialogue-based English CALL system. Experiments show that a speech recognizer achieved higher recognition accuracy with the reduced phoneme set than with the canonical phoneme set.},
keywords={},
doi={10.1587/transinf.2014EDP7168},
ISSN={1745-1361},
month={January},}

Copy

TY - JOUR
TI - Phoneme Set Design for Speech Recognition of English by Japanese
T2 - IEICE TRANSACTIONS on Information
SP - 148
EP - 156
AU - Xiaoyun WANG
AU - Jinsong ZHANG
AU - Masafumi NISHIDA
AU - Seiichi YAMAMOTO
PY - 2015
DO - 10.1587/transinf.2014EDP7168
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2015
AB - This paper describes a novel method to improve the performance of second language speech recognition when the mother tongue of users is known. Considering that second language speech usually includes less fluent pronunciation and more frequent pronunciation mistakes, the authors propose using a reduced phoneme set generated by a phonetic decision tree (PDT)-based top-down sequential splitting method instead of the canonical one of the second language. The authors verify the efficacy of the proposed method using second language speech collected with a translation game type dialogue-based English CALL system. Experiments show that a speech recognizer achieved higher recognition accuracy with the reduced phoneme set than with the canonical phoneme set.
ER -