Publication Details

Discriminative Classifiers for Phonotactic Language Recognition with iVectors

SOUFIFAR Mehdi Mohammad, CUMANI Sandro, BURGET Lukáš and ČERNOCKÝ Jan. Discriminative Classifiers for Phonotactic Language Recognition with iVectors. In: Proc. International Conference on Acoustics, Speech, and Signal Processing 2012. Kyoto: IEEE Signal Processing Society, 2012, pp. 4853-4856. ISBN 978-1-4673-0044-5.
Czech title
Diskriminativní klasifikátory pro fonotaktické rozpoznávání jazyka s i-vektory
Type
conference paper
Language
english
Authors
Soufifar Mehdi Mohammad, Ing. (FIT BUT)
Cumani Sandro (POLITO)
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT)
Černocký Jan, prof. Dr. Ing. (DCGM FIT BUT)
URL
Keywords

Phonotactic iVector, Discriminative classifier, Support vector machine, Logistic regression

Abstract

The paper is about phonotactic models based on bags of n-grams representations and discriminative classifiers are a popular approach to the language recognition problem.

Annotation

Phonotactic models based on bags of n-grams representations and discriminative classifiers are a popular approach to the language recognition problem. However, the large size of n-gram count vectors brings about some difficulties in discriminative classifiers. The subspace Multinomial model was recently proposed to effectively represent information contained in the n-grams using low-dimensional iVectors. The availability of a low-dimensional feature vector allows investigating different post-processing techniques and different classifiers to improve recognition performance. In this work, we analyze a set of discriminative classifiers based on Support Vector Machines and Logistic Regression and we propose an iVector post-processing technique which allows to improve recognition performance. The proposed systems are evaluated on the NIST LRE 2009 task.

Published
2012
Pages
4853-4856
Proceedings
Proc. International Conference on Acoustics, Speech, and Signal Processing 2012
Conference
The 37th International Conference on Acoustics, Speech, and Signal Processing, Kyoto, JP
ISBN
978-1-4673-0044-5
Publisher
IEEE Signal Processing Society
Place
Kyoto, JP
DOI
BibTeX
@INPROCEEDINGS{FITPUB9909,
   author = "Mohammad Mehdi Soufifar and Sandro Cumani and Luk\'{a}\v{s} Burget and Jan \v{C}ernock\'{y}",
   title = "Discriminative Classifiers for Phonotactic Language Recognition with iVectors",
   pages = "4853--4856",
   booktitle = "Proc. International Conference on Acoustics, Speech, and Signal Processing 2012",
   year = 2012,
   location = "Kyoto, JP",
   publisher = "IEEE Signal Processing Society",
   ISBN = "978-1-4673-0044-5",
   doi = "10.1109/ICASSP.2012.6289006",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/9909"
}
Back to top