Sciweavers

ICASSP
2011
IEEE
12 years 7 months ago
Subspace constrained LU decomposition of FMLLR for rapid adaptation
This paper describes subspace constrained feature space maximum likelihood linear regression (FMLLR) for rapid adaptation. The test speaker’s FMLLR rotation matrix is decomposed...
Lei Jia, Dong Yu, Bo Xu
ICASSP
2011
IEEE
12 years 7 months ago
Rapid feature space MLLR speaker adaptation with bilinear models
In this paper, we propose a novel method for rapid feature space Maximum Likelihood Linear Regression (FMLLR) speaker adaptation based on bilinear models. When the amount of adapt...
Shilei Zhang, Peder A. Olsen, Yong Qin
ICASSP
2011
IEEE
12 years 7 months ago
A sampling-based environment population projection approach for rapid acoustic model adaptation
We propose an environment population projection (EPP) approach for rapid acoustic model adaptation to reduce environment mismatches with limited amounts of adaptation data. This a...
Yu Tsao, Shigeki Matsuda, Shinsuke Sakai, Ryosuke ...
INTERSPEECH
2010
12 years 10 months ago
Adaptation of a tongue shape model by local feature transformations
Reconstructing the full contour of the tongue from the position of 3 to 4 landmarks on it is useful in articulatory speech work. This can be done with submillimetric accuracy usin...
Chao Qin, Miguel Á. Carreira-Perpiñ&...
INTERSPEECH
2010
12 years 10 months ago
Discriminative adaptation based on fast combination of DMAP and dfMLLR
This paper investigates the combination of discriminative adaptation techniques. The discriminative Maximum A-Posteriori (DMAP) adaptation and discriminative feature Maximum Likel...
Lukás Machlica, Zbynek Zajíc, Ludek ...
INTERSPEECH
2010
12 years 10 months ago
Augmentation of adaptation data
Linear regression based speaker adaptation approaches can improve Automatic Speech Recognition (ASR) accuracy significantly for a target speaker. However, when the available adapt...
Ravichander Vipperla, Steve Renals, Joe Frankel
CVPR
2011
IEEE
12 years 11 months ago
Adapting an Object Detector by Considering the Worst Case: a Conservative Approach
The performance of an offline-trained classifier can be improved on-site by adapting the classifier towards newly acquired data. However, the adaptation rate is a tuning parame...
Guang Chen, TonyX. Han
ICASSP
2008
IEEE
13 years 10 months ago
Maximum conditional likelihood linear regression and maximum a posteriori for hidden conditional random fields speaker adaptatio
This paper shows how to improve Hidden Conditional Random Fields (HCRFs) for phone classification by applying various speaker adaptation techniques. These include Maximum A Poste...
Yun-Hsuan Sung, Constantinos Boulis, Daniel Jurafs...