Intra-cluster training strategy for deep learning with applications to language identification

Alan Joseph Bekker; Irit Opher; Itsik Lapidot; Jacob Goldberger

doi:10.1109/mlsp.2016.7738896

Intra-cluster training strategy for deep learning with applications to language identification

Alan Joseph Bekker
, Irit Opher
, Itsik Lapidot
, Jacob Goldberger

Bar-Ilan University - The Alexander Kofkin Faculty of Engineering

Afeka Acad. Coll. of Eng

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

14 Scopus citations

Abstract

In this study we address the problem of training a neural network for language identification using speech samples in the form of i-vectors. Our approach involves training a classifier and analyzing the obtained confusion matrix. We cluster the languages by simultaneously clustering the columns and the rows of the confusion matrix. The language clusters are then used to define a modified cost function for training a neural-network that focuses on distinguishing between the true language and languages within the same cluster. The results show enhanced language identification on the NIST 2015 language identification dataset.

Original language	English
Title of host publication	2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016
Editors	Francesco A. N. Palmieri, Aurelio Uncini, Kostas Diamantaras, Jan Larsen
Publisher	IEEE Computer Society
ISBN (Electronic)	9781509007462
DOIs	https://doi.org/10.1109/mlsp.2016.7738896
State	Published - 8 Nov 2016
Event	26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Vietri sul Mare, Salerno, Italy Duration: 13 Sep 2016 → 16 Sep 2016

Publication series

Name	IEEE International Workshop on Machine Learning for Signal Processing, MLSP
Volume	2016-November
ISSN (Print)	2161-0363
ISSN (Electronic)	2161-0371

Conference

Conference	26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016
Country/Territory	Italy
City	Vietri sul Mare, Salerno
Period	13/09/16 → 16/09/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Keywords

Confusion matrix
clustering
language identification

Access to Document

10.1109/mlsp.2016.7738896

Cite this

Bekker, A. J., Opher, I., Lapidot, I., & Goldberger, J. (2016). Intra-cluster training strategy for deep learning with applications to language identification. In F. A. N. Palmieri, A. Uncini, K. Diamantaras, & J. Larsen (Eds.), 2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016 Article 7738896 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2016-November). IEEE Computer Society. https://doi.org/10.1109/mlsp.2016.7738896

Bekker, Alan Joseph ; Opher, Irit ; Lapidot, Itsik et al. / Intra-cluster training strategy for deep learning with applications to language identification. 2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016. editor / Francesco A. N. Palmieri ; Aurelio Uncini ; Kostas Diamantaras ; Jan Larsen. IEEE Computer Society, 2016. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP).

@inproceedings{bb173de9e36f484fa42076dc974059ac,

title = "Intra-cluster training strategy for deep learning with applications to language identification",

abstract = "In this study we address the problem of training a neural network for language identification using speech samples in the form of i-vectors. Our approach involves training a classifier and analyzing the obtained confusion matrix. We cluster the languages by simultaneously clustering the columns and the rows of the confusion matrix. The language clusters are then used to define a modified cost function for training a neural-network that focuses on distinguishing between the true language and languages within the same cluster. The results show enhanced language identification on the NIST 2015 language identification dataset.",

keywords = "Confusion matrix, clustering, language identification",

author = "Bekker, \{Alan Joseph\} and Irit Opher and Itsik Lapidot and Jacob Goldberger",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 ; Conference date: 13-09-2016 Through 16-09-2016",

year = "2016",

month = nov,

day = "8",

doi = "10.1109/mlsp.2016.7738896",

language = "אנגלית",

series = "IEEE International Workshop on Machine Learning for Signal Processing, MLSP",

publisher = "IEEE Computer Society",

editor = "Palmieri, \{Francesco A. N.\} and Aurelio Uncini and Kostas Diamantaras and Jan Larsen",

booktitle = "2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016",

address = "ארצות הברית",

}

Bekker, AJ, Opher, I, Lapidot, I & Goldberger, J 2016, Intra-cluster training strategy for deep learning with applications to language identification. in FAN Palmieri, A Uncini, K Diamantaras & J Larsen (eds), 2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016., 7738896, IEEE International Workshop on Machine Learning for Signal Processing, MLSP, vol. 2016-November, IEEE Computer Society, 26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016, Vietri sul Mare, Salerno, Italy, 13/09/16. https://doi.org/10.1109/mlsp.2016.7738896

Intra-cluster training strategy for deep learning with applications to language identification. / Bekker, Alan Joseph; Opher, Irit; Lapidot, Itsik et al.
2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016. ed. / Francesco A. N. Palmieri; Aurelio Uncini; Kostas Diamantaras; Jan Larsen. IEEE Computer Society, 2016. 7738896 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2016-November).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Intra-cluster training strategy for deep learning with applications to language identification

AU - Bekker, Alan Joseph

AU - Opher, Irit

AU - Lapidot, Itsik

AU - Goldberger, Jacob

PY - 2016/11/8

Y1 - 2016/11/8

N2 - In this study we address the problem of training a neural network for language identification using speech samples in the form of i-vectors. Our approach involves training a classifier and analyzing the obtained confusion matrix. We cluster the languages by simultaneously clustering the columns and the rows of the confusion matrix. The language clusters are then used to define a modified cost function for training a neural-network that focuses on distinguishing between the true language and languages within the same cluster. The results show enhanced language identification on the NIST 2015 language identification dataset.

AB - In this study we address the problem of training a neural network for language identification using speech samples in the form of i-vectors. Our approach involves training a classifier and analyzing the obtained confusion matrix. We cluster the languages by simultaneously clustering the columns and the rows of the confusion matrix. The language clusters are then used to define a modified cost function for training a neural-network that focuses on distinguishing between the true language and languages within the same cluster. The results show enhanced language identification on the NIST 2015 language identification dataset.

KW - Confusion matrix

KW - clustering

KW - language identification

UR - https://www.scopus.com/pages/publications/85002489698

U2 - 10.1109/mlsp.2016.7738896

DO - 10.1109/mlsp.2016.7738896

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85002489698

T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP

BT - 2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016

A2 - Palmieri, Francesco A. N.

A2 - Uncini, Aurelio

A2 - Diamantaras, Kostas

A2 - Larsen, Jan

PB - IEEE Computer Society

T2 - 26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016

Y2 - 13 September 2016 through 16 September 2016

ER -

Bekker AJ, Opher I, Lapidot I, Goldberger J. Intra-cluster training strategy for deep learning with applications to language identification. In Palmieri FAN, Uncini A, Diamantaras K, Larsen J, editors, 2016 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2016. IEEE Computer Society. 2016. 7738896. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP). doi: 10.1109/mlsp.2016.7738896

Intra-cluster training strategy for deep learning with applications to language identification

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this