Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English

Roland Goecke; J Millar

Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English

Roland Goecke, J Millar

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

Abstract

After decades of research, automatic speech processing has become more and more viable in recent years. Audio-video speech recognition has been shown to improve the recognition rate in noise-degraded environments. However, which audio and video speech parameters to choose for an optimal system and how they are related is still an open research issue. Here we present a number of statistical analyses that aim at increasing our understanding of such audio-video relationships. In particular, we look at the canonical correlation analysis and the coinertia analysis which investigate the relationship of linear combinations of parameters. The analyses are performed on Australian English as an example.

Original language	English
Title of host publication	AVSP 2003: ISCA Tutorial and Research workshop on audio visual speech processing
Editors	Jean-Luc Schwartz, David Sodoyer
Place of Publication	Canada
Publisher	ISCA
Pages	133-138
Number of pages	6
Publication status	Published - 2003
Event	AVSP 2003 - St Jorioz, France Duration: 4 Sept 2003 → 7 Sept 2003

Conference

Conference	AVSP 2003
Country/Territory	France
City	St Jorioz
Period	4/09/03 → 7/09/03

Access to Document

https://www.isca-speech.org/archive_open/archive_papers/avsp03/av03_133.pdf

Cite this

@inproceedings{6cd81aa64a824cdfabb0ade0740e8ef8,

title = "Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English",

abstract = "After decades of research, automatic speech processing has become more and more viable in recent years. Audio-video speech recognition has been shown to improve the recognition rate in noise-degraded environments. However, which audio and video speech parameters to choose for an optimal system and how they are related is still an open research issue. Here we present a number of statistical analyses that aim at increasing our understanding of such audio-video relationships. In particular, we look at the canonical correlation analysis and the coinertia analysis which investigate the relationship of linear combinations of parameters. The analyses are performed on Australian English as an example.",

author = "Roland Goecke and J Millar",

year = "2003",

language = "English",

pages = "133--138",

editor = "Jean-Luc Schwartz and David Sodoyer",

booktitle = "AVSP 2003: ISCA Tutorial and Research workshop on audio visual speech processing",

publisher = "ISCA",

note = "AVSP 2003 ; Conference date: 04-09-2003 Through 07-09-2003",

}

Goecke, R & Millar, J 2003, Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English. in J-L Schwartz & D Sodoyer (eds), AVSP 2003: ISCA Tutorial and Research workshop on audio visual speech processing. ISCA, Canada, pp. 133-138, AVSP 2003, St Jorioz, France, 4/09/03. <https://www.isca-speech.org/archive_open/archive_papers/avsp03/av03_133.pdf>

Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English. / Goecke, Roland; Millar, J.
AVSP 2003: ISCA Tutorial and Research workshop on audio visual speech processing. ed. / Jean-Luc Schwartz; David Sodoyer. Canada: ISCA, 2003. p. 133-138.

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

TY - GEN

T1 - Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English

AU - Goecke, Roland

AU - Millar, J

PY - 2003

Y1 - 2003

N2 - After decades of research, automatic speech processing has become more and more viable in recent years. Audio-video speech recognition has been shown to improve the recognition rate in noise-degraded environments. However, which audio and video speech parameters to choose for an optimal system and how they are related is still an open research issue. Here we present a number of statistical analyses that aim at increasing our understanding of such audio-video relationships. In particular, we look at the canonical correlation analysis and the coinertia analysis which investigate the relationship of linear combinations of parameters. The analyses are performed on Australian English as an example.

AB - After decades of research, automatic speech processing has become more and more viable in recent years. Audio-video speech recognition has been shown to improve the recognition rate in noise-degraded environments. However, which audio and video speech parameters to choose for an optimal system and how they are related is still an open research issue. Here we present a number of statistical analyses that aim at increasing our understanding of such audio-video relationships. In particular, we look at the canonical correlation analysis and the coinertia analysis which investigate the relationship of linear combinations of parameters. The analyses are performed on Australian English as an example.

M3 - Conference contribution

SP - 133

EP - 138

BT - AVSP 2003: ISCA Tutorial and Research workshop on audio visual speech processing

A2 - Schwartz, Jean-Luc

A2 - Sodoyer, David

PB - ISCA

CY - Canada

T2 - AVSP 2003

Y2 - 4 September 2003 through 7 September 2003

ER -

Statistical Analysis of the Relationship between Audio and Video speech parameters for Australian English

Abstract

Conference

Access to Document

Fingerprint

Cite this