Graduate School and Research Center in Digital Sciences

A simplified 2-layer text-dependent speaker authentication system

Valenti, Giacomo; Daniel, Adrien; Evans, Nicholas

AES 2017, 143rd International AES Convention, October 18-21, 2017, New York, USA

This paper describes a variation of the well-known HiLAM approach to speaker authentication which enables reliable text-dependent speaker recognition with short-duration enrollment. The modifications introduced in this system eliminate the need for an intermediate text-independent speaker model. While the simplified system is admittedly a modest modification to the original work, it delivers comparable levels of automatic speaker verification performance while requiring 97% less speaker enrollment data. Such a significant reduction in enrollment data improves usability and supports speaker authentication for smart device and Internet of Things applications.

Document Bibtex

Title:A simplified 2-layer text-dependent speaker authentication system
Type:Conference
Language:English
City:New York
Country:UNITED STATES
Date:
Department:Digital Security
Eurecom ref:5323
Copyright: Copyright AES. Personal use of this material is permitted. The definitive version of this paper was published in AES 2017, 143rd International AES Convention, October 18-21, 2017, New York, USA and is available at :
Bibtex: @inproceedings{EURECOM+5323, year = {2017}, title = {{A} simplified 2-layer text-dependent speaker authentication system}, author = {{V}alenti, {G}iacomo and {D}aniel, {A}drien and {E}vans, {N}icholas}, booktitle = {{AES} 2017, 143rd {I}nternational {AES} {C}onvention, {O}ctober 18-21, 2017, {N}ew {Y}ork, {USA}}, address = {{N}ew {Y}ork, {UNITED} {STATES}}, month = {10}, url = {http://www.eurecom.fr/publication/5323} }
See also: