MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion

Narjes Bozorg; Michael T. Johnson

doi:10.1109/ISSPIT47144.2019.9001752

MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion

Narjes Bozorg, Michael T. Johnson

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This paper presents an improved method for kinematic-independent acoustic-to-articulatory inversion, using acoustic adaptation to estimate weights for articulatory model creation from reference speakers. Paired acoustic and articulatory data from the Marquette Electromagnetic Articulography corpus of Mandarin Accented English (EMAMAE) are used for experimental evaluation. The new method is a modification of the Parallel Reference Speaker Weighting (PRSW) inversion algorithm, in which two separate methods are used for acoustic and articulatory adaptation. A Maximum Likelihood Linear Regression (MLLR) approach is used for acoustic adaptation model and the PRSW weighted reference speaker approach is used for articulatory model adaptation. The new MLLR-PRSW adaptation method outperforms the baseline PRSW method on inversion of new test subjects where no kinematic data is used for training, providing estimated trajectories very close to the results from speaker dependent models that do use kinematic data.

Original language	English
Title of host publication	2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019
ISBN (Electronic)	9781728153414
DOIs	https://doi.org/10.1109/ISSPIT47144.2019.9001752
State	Published - Dec 2019
Event	19th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2019 - Ajman, United Arab Emirates Duration: Dec 10 2019 → Dec 12 2019

Publication series

Name	2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019

Conference

Conference	19th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2019
Country/Territory	United Arab Emirates
City	Ajman
Period	12/10/19 → 12/12/19

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Keywords

Speaker Independent acoustic-to-articulatory inversion
electromagnetic articulography
maximum likelihood linear regression
parallel reference speaker weighting

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems
Signal Processing
Information Systems and Management

Access to Document

10.1109/ISSPIT47144.2019.9001752

Cite this

Bozorg, N., & Johnson, M. T. (2019). MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion. In 2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019 [9001752] (2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019). https://doi.org/10.1109/ISSPIT47144.2019.9001752

@inproceedings{4cf9ab568970462ea4c75529b96bf211,

title = "MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion",

abstract = "This paper presents an improved method for kinematic-independent acoustic-to-articulatory inversion, using acoustic adaptation to estimate weights for articulatory model creation from reference speakers. Paired acoustic and articulatory data from the Marquette Electromagnetic Articulography corpus of Mandarin Accented English (EMAMAE) are used for experimental evaluation. The new method is a modification of the Parallel Reference Speaker Weighting (PRSW) inversion algorithm, in which two separate methods are used for acoustic and articulatory adaptation. A Maximum Likelihood Linear Regression (MLLR) approach is used for acoustic adaptation model and the PRSW weighted reference speaker approach is used for articulatory model adaptation. The new MLLR-PRSW adaptation method outperforms the baseline PRSW method on inversion of new test subjects where no kinematic data is used for training, providing estimated trajectories very close to the results from speaker dependent models that do use kinematic data.",

keywords = "Speaker Independent acoustic-to-articulatory inversion, electromagnetic articulography, maximum likelihood linear regression, parallel reference speaker weighting",

author = "Narjes Bozorg and Johnson, {Michael T.}",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; null ; Conference date: 10-12-2019 Through 12-12-2019",

year = "2019",

month = dec,

doi = "10.1109/ISSPIT47144.2019.9001752",

language = "English",

series = "2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019",

booktitle = "2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019",

}

Bozorg, N & Johnson, MT 2019, MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion. in 2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019., 9001752, 2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019, 19th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2019, Ajman, United Arab Emirates, 12/10/19. https://doi.org/10.1109/ISSPIT47144.2019.9001752

MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion. / Bozorg, Narjes; Johnson, Michael T.

2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019. 2019. 9001752 (2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion

AU - Bozorg, Narjes

AU - Johnson, Michael T.

PY - 2019/12

Y1 - 2019/12

N2 - This paper presents an improved method for kinematic-independent acoustic-to-articulatory inversion, using acoustic adaptation to estimate weights for articulatory model creation from reference speakers. Paired acoustic and articulatory data from the Marquette Electromagnetic Articulography corpus of Mandarin Accented English (EMAMAE) are used for experimental evaluation. The new method is a modification of the Parallel Reference Speaker Weighting (PRSW) inversion algorithm, in which two separate methods are used for acoustic and articulatory adaptation. A Maximum Likelihood Linear Regression (MLLR) approach is used for acoustic adaptation model and the PRSW weighted reference speaker approach is used for articulatory model adaptation. The new MLLR-PRSW adaptation method outperforms the baseline PRSW method on inversion of new test subjects where no kinematic data is used for training, providing estimated trajectories very close to the results from speaker dependent models that do use kinematic data.

AB - This paper presents an improved method for kinematic-independent acoustic-to-articulatory inversion, using acoustic adaptation to estimate weights for articulatory model creation from reference speakers. Paired acoustic and articulatory data from the Marquette Electromagnetic Articulography corpus of Mandarin Accented English (EMAMAE) are used for experimental evaluation. The new method is a modification of the Parallel Reference Speaker Weighting (PRSW) inversion algorithm, in which two separate methods are used for acoustic and articulatory adaptation. A Maximum Likelihood Linear Regression (MLLR) approach is used for acoustic adaptation model and the PRSW weighted reference speaker approach is used for articulatory model adaptation. The new MLLR-PRSW adaptation method outperforms the baseline PRSW method on inversion of new test subjects where no kinematic data is used for training, providing estimated trajectories very close to the results from speaker dependent models that do use kinematic data.

KW - Speaker Independent acoustic-to-articulatory inversion

KW - electromagnetic articulography

KW - maximum likelihood linear regression

KW - parallel reference speaker weighting

UR - http://www.scopus.com/inward/record.url?scp=85081296677&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85081296677&partnerID=8YFLogxK

U2 - 10.1109/ISSPIT47144.2019.9001752

DO - 10.1109/ISSPIT47144.2019.9001752

M3 - Conference contribution

AN - SCOPUS:85081296677

T3 - 2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019

BT - 2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019

Y2 - 10 December 2019 through 12 December 2019

ER -

MLLR-PRSW for Kinematic-Independent Acoustic-to-Articulatory Inversion

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this