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

facebooktwittermailshare

Sufficiency quantification for seamless text-independent speaker enrollment

Abstract: 

Text-independent speaker recognition (TI-SR) requires a lengthy enrollment process that involves asking dedicated time from the user to create a reliable model of their voice. Seamless enrollment is a highly attractive feature which refers to the enrollment process that happens in the background and asks for no dedicated time from the user. One of the key problems in a fully automated seamless enrollment process is to determine the sufficiency of a given utterance collection for the purpose of TI-SR. No known metric exists in the literature to quantify sufficiency. This paper introduces a novel metric called phoneme-richness score. Quality of a sufficiency metric can be assessed via its correlation with the TI-SR performance. Our assessment shows that phoneme-richness score achieves -0.96 correlation with TI-SR performance (measured in equal error rate), which is highly significant, whereas a naive sufficiency metric like speech duration achieves only -0.68 correlation.

up
0 users have voted:

Paper Details

Authors:
Gokcen Cilingir, Jonathan Huang, Mandar S Joshi, Narayan Biswal
Submitted On:
13 July 2018 - 3:38pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Gokcen Cilingir
Paper Code:
SP-P5.8
Document Year:
2018
Cite

Document Files

Poster presented at ICASSP 2018

(45 downloads)

Paper for ICASSP 2018

(38 downloads)

Subscribe

[1] Gokcen Cilingir, Jonathan Huang, Mandar S Joshi, Narayan Biswal, "Sufficiency quantification for seamless text-independent speaker enrollment", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3379. Accessed: Sep. 25, 2018.
@article{3379-18,
url = {http://sigport.org/3379},
author = {Gokcen Cilingir; Jonathan Huang; Mandar S Joshi; Narayan Biswal },
publisher = {IEEE SigPort},
title = {Sufficiency quantification for seamless text-independent speaker enrollment},
year = {2018} }
TY - EJOUR
T1 - Sufficiency quantification for seamless text-independent speaker enrollment
AU - Gokcen Cilingir; Jonathan Huang; Mandar S Joshi; Narayan Biswal
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3379
ER -
Gokcen Cilingir, Jonathan Huang, Mandar S Joshi, Narayan Biswal. (2018). Sufficiency quantification for seamless text-independent speaker enrollment. IEEE SigPort. http://sigport.org/3379
Gokcen Cilingir, Jonathan Huang, Mandar S Joshi, Narayan Biswal, 2018. Sufficiency quantification for seamless text-independent speaker enrollment. Available at: http://sigport.org/3379.
Gokcen Cilingir, Jonathan Huang, Mandar S Joshi, Narayan Biswal. (2018). "Sufficiency quantification for seamless text-independent speaker enrollment." Web.
1. Gokcen Cilingir, Jonathan Huang, Mandar S Joshi, Narayan Biswal. Sufficiency quantification for seamless text-independent speaker enrollment [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3379