Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification

Abstract: 

This paper presents supervised feature learning approaches for speaker identification that rely on nonnegative matrix factorisation. Recent studies have shown that group nonnegative matrix factorisation and task-driven supervised dictionary learning can help performing effective feature learning for audio classification problems.

This paper proposes to integrate a recent method that relies on group nonnegative matrix factorisation into a task-driven supervised framework for speaker identification. The goal is to capture both the speaker variability and the session variability while exploiting the
discriminative learning aspect of the task-driven approach. Results on a subset of the ESTER corpus prove that the proposed approach can be competitive with I-vectors.

Code available at https://github.com/rserizel/TGNMF

up
0 users have voted:

Paper Details

Authors:
victor bisot, slim essid, gaël richard
Submitted On:
1 March 2017 - 4:34am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Romain Serizel
Paper Code:
AASP-L2.2
Document Year:
2017
Cite

Document Files

Slide for the presentation

(34 downloads)

Subscribe

[1] victor bisot, slim essid, gaël richard, "Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1539. Accessed: Jun. 28, 2017.
@article{1539-17,
url = {http://sigport.org/1539},
author = {victor bisot; slim essid; gaël richard },
publisher = {IEEE SigPort},
title = {Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification},
year = {2017} }
TY - EJOUR
T1 - Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification
AU - victor bisot; slim essid; gaël richard
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1539
ER -
victor bisot, slim essid, gaël richard. (2017). Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification. IEEE SigPort. http://sigport.org/1539
victor bisot, slim essid, gaël richard, 2017. Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification. Available at: http://sigport.org/1539.
victor bisot, slim essid, gaël richard. (2017). "Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification." Web.
1. victor bisot, slim essid, gaël richard. Supervised group nonnegative matrix factorisation with similarity constraints and applications to speaker identification [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1539