Speaker Adaptation Based on PARAFAC2 of Transformation Matrices for Continuous Speech Recognition

Yongwon JEONG; Sangjun LIM; Young Kuk KIM; Hyung Soon KIM

doi:10.1587/transinf.E96.D.2152

Speaker Adaptation Based on PARAFAC2 of Transformation Matrices for Continuous Speech Recognition

Yongwon JEONG, Sangjun LIM, Young Kuk KIM, Hyung Soon KIM

Full Text Views

0

Share
Cite this

Summary :

We present an acoustic model adaptation method where the transformation matrix for a new speaker is given by the product of bases and a weight matrix. The bases are built from the parallel factor analysis 2 (PARAFAC2) of training speakers' transformation matrices. We perform continuous speech recognition experiments using the WSJ0 corpus.

Publication: IEICE TRANSACTIONS on Information Vol.E96-D No.9 pp.2152-2155

Publication Date: 2013/09/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E96.D.2152

Type of Manuscript: LETTER

Category: Speech and Hearing

Authors

Yongwon JEONG
  Pusan National University
Sangjun LIM
  Pusan National University
Young Kuk KIM
  AirPlug
Hyung Soon KIM
  Pusan National University

Keyword

maximum likelihood linear regression, parallel factor analysis, PARAFAC2, speaker adaptation, speech recognition

Cite this

Copy

Yongwon JEONG, Sangjun LIM, Young Kuk KIM, Hyung Soon KIM, "Speaker Adaptation Based on PARAFAC2 of Transformation Matrices for Continuous Speech Recognition" in IEICE TRANSACTIONS on Information, vol. E96-D, no. 9, pp. 2152-2155, September 2013, doi: 10.1587/transinf.E96.D.2152.
Abstract: We present an acoustic model adaptation method where the transformation matrix for a new speaker is given by the product of bases and a weight matrix. The bases are built from the parallel factor analysis 2 (PARAFAC2) of training speakers' transformation matrices. We perform continuous speech recognition experiments using the WSJ0 corpus.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E96.D.2152/_p

Copy

@ARTICLE{e96-d_9_2152,
author={Yongwon JEONG, Sangjun LIM, Young Kuk KIM, Hyung Soon KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Speaker Adaptation Based on PARAFAC2 of Transformation Matrices for Continuous Speech Recognition},
year={2013},
volume={E96-D},
number={9},
pages={2152-2155},
abstract={We present an acoustic model adaptation method where the transformation matrix for a new speaker is given by the product of bases and a weight matrix. The bases are built from the parallel factor analysis 2 (PARAFAC2) of training speakers' transformation matrices. We perform continuous speech recognition experiments using the WSJ0 corpus.},
keywords={},
doi={10.1587/transinf.E96.D.2152},
ISSN={1745-1361},
month={September},}

Copy

TY - JOUR
TI - Speaker Adaptation Based on PARAFAC2 of Transformation Matrices for Continuous Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 2152
EP - 2155
AU - Yongwon JEONG
AU - Sangjun LIM
AU - Young Kuk KIM
AU - Hyung Soon KIM
PY - 2013
DO - 10.1587/transinf.E96.D.2152
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2013
AB - We present an acoustic model adaptation method where the transformation matrix for a new speaker is given by the product of bases and a weight matrix. The bases are built from the parallel factor analysis 2 (PARAFAC2) of training speakers' transformation matrices. We perform continuous speech recognition experiments using the WSJ0 corpus.
ER -