Information geometry of topology preserving adaptation

Mustafa (Kemal) Sonmez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We consider adaptation by topologically smooth transformation with applications to environment and speaker adaptation for robust speech recognition. Specifically, the trade-off between global affine transformations that fail to capture local variation but preserve topology and local class dependent bias transformations that have more resolution but may destroy the topology of the reference model is addressed. We cast the problem of topology preservation of the reference model in an information divergence geometry framework and derive a class of alternating minimization algorithms that aims to preserve topology explicitly during adaptation.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
PublisherIEEE
Pages3743-3746
Number of pages4
Volume6
StatePublished - 2000
Externally publishedYes
Event2000 IEEE Interntional Conference on Acoustics, Speech, and Signal Processing - Istanbul, Turkey
Duration: Jun 5 2000Jun 9 2000

Other

Other2000 IEEE Interntional Conference on Acoustics, Speech, and Signal Processing
CityIstanbul, Turkey
Period6/5/006/9/00

Fingerprint

preserving
topology
Topology
Geometry
geometry
speech recognition
Speech recognition
casts
divergence
optimization

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Sonmez, M. K. (2000). Information geometry of topology preserving adaptation. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 6, pp. 3743-3746). IEEE.

Information geometry of topology preserving adaptation. / Sonmez, Mustafa (Kemal).

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 6 IEEE, 2000. p. 3743-3746.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Sonmez, MK 2000, Information geometry of topology preserving adaptation. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 6, IEEE, pp. 3743-3746, 2000 IEEE Interntional Conference on Acoustics, Speech, and Signal Processing, Istanbul, Turkey, 6/5/00.
Sonmez MK. Information geometry of topology preserving adaptation. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 6. IEEE. 2000. p. 3743-3746
Sonmez, Mustafa (Kemal). / Information geometry of topology preserving adaptation. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 6 IEEE, 2000. pp. 3743-3746
@inproceedings{78d0003df87b417b9d599feeb0be8d18,
title = "Information geometry of topology preserving adaptation",
abstract = "We consider adaptation by topologically smooth transformation with applications to environment and speaker adaptation for robust speech recognition. Specifically, the trade-off between global affine transformations that fail to capture local variation but preserve topology and local class dependent bias transformations that have more resolution but may destroy the topology of the reference model is addressed. We cast the problem of topology preservation of the reference model in an information divergence geometry framework and derive a class of alternating minimization algorithms that aims to preserve topology explicitly during adaptation.",
author = "Sonmez, {Mustafa (Kemal)}",
year = "2000",
language = "English (US)",
volume = "6",
pages = "3743--3746",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Information geometry of topology preserving adaptation

AU - Sonmez, Mustafa (Kemal)

PY - 2000

Y1 - 2000

N2 - We consider adaptation by topologically smooth transformation with applications to environment and speaker adaptation for robust speech recognition. Specifically, the trade-off between global affine transformations that fail to capture local variation but preserve topology and local class dependent bias transformations that have more resolution but may destroy the topology of the reference model is addressed. We cast the problem of topology preservation of the reference model in an information divergence geometry framework and derive a class of alternating minimization algorithms that aims to preserve topology explicitly during adaptation.

AB - We consider adaptation by topologically smooth transformation with applications to environment and speaker adaptation for robust speech recognition. Specifically, the trade-off between global affine transformations that fail to capture local variation but preserve topology and local class dependent bias transformations that have more resolution but may destroy the topology of the reference model is addressed. We cast the problem of topology preservation of the reference model in an information divergence geometry framework and derive a class of alternating minimization algorithms that aims to preserve topology explicitly during adaptation.

UR - http://www.scopus.com/inward/record.url?scp=0033678499&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033678499&partnerID=8YFLogxK

M3 - Conference contribution

VL - 6

SP - 3743

EP - 3746

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

PB - IEEE

ER -