Publication Details

iVector-Based Discriminative Adaptation for Automatic Speech Recognition

KARAFIÁT Martin, BURGET Lukáš, MATĚJKA Pavel, GLEMBEK Ondřej and ČERNOCKÝ Jan. iVector-Based Discriminative Adaptation for Automatic Speech Recognition. In: Proceedings of ASRU 2011. Hilton Waikoloa Village, Big Island, Hawaii: IEEE Signal Processing Society, 2011, pp. 152-157. ISBN 978-1-4673-0366-8.
Czech title
Diskriminativní adaptace pro automatické rozpoznávání řeči založená na i-vektorech
Type
conference paper
Language
english
Authors
Karafiát Martin, Ing., Ph.D. (DCGM FIT BUT)
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT)
Matějka Pavel, Ing., Ph.D. (DCGM FIT BUT)
Glembek Ondřej, Ing., Ph.D. (DCGM FIT BUT)
Černocký Jan, prof. Dr. Ing. (DCGM FIT BUT)
URL
Keywords

Automatic speech recognition, I-vector, Discriminative adaptation

Abstract

The iVector is a low-dimensional fixed-length representation of information about speaker and acoustic environment. To utilize iVectors for adaptation, region dependent linear transforms (RDLT) are discriminatively trained using the MPE criterion on large amounts of annotated data to extract the relevant information from iVectors and to compensate speech features. The approach was tested on standard CTS data. We found it to be complementary to common adaptation techniques. On a well-tuned RDLT system with standard CMLLR adaptation we reached an 0.8% additive absolute WER improvement.

Annotation

This work describes a novel technique for discriminative feature-level adaptation for automatic speech recognition. The concept of iVectors popular in speaker recognition is used to extract information about a speaker or acoustic environment from a speech segment. The iVector is a low-dimensional fixed-length representation of such information. To utilize iVectors for adaptation, region dependent linear transforms (RDLT) are discriminatively trained using the MPE criterion on large amounts of annotated data to extract the relevant information from iVectors and to compensate speech features. The approach was tested on standard CTS data. We found it to be complementary to common adaptation techniques. On a well-tuned RDLT system with standard CMLLR adaptation we reached an 0.8% additive absolute WER improvement.

Published
2011
Pages
152-157
Proceedings
Proceedings of ASRU 2011
Conference
IEEE 2011 Workshop on Automatic Speech Recognition and Understanding, Hilton Waikoloa Village Resort, Big Island, Hawaii, US
ISBN
978-1-4673-0366-8
Publisher
IEEE Signal Processing Society
Place
Hilton Waikoloa Village, Big Island, Hawaii, US
BibTeX 
@INPROCEEDINGS{FITPUB9762,
   author = "Martin Karafi\'{a}t and Luk\'{a}\v{s} Burget and Pavel Mat\v{e}jka and Ond\v{r}ej Glembek and Jan \v{C}ernock\'{y}",
   title = "iVector-Based Discriminative Adaptation for Automatic Speech Recognition",
   pages = "152--157",
   booktitle = "Proceedings of ASRU 2011",
   year = 2011,
   location = "Hilton Waikoloa Village, Big Island, Hawaii, US",
   publisher = "IEEE Signal Processing Society",
   ISBN = "978-1-4673-0366-8",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/9762"
}
Back to top