Conference paper

CUMANI Sandro, GLEMBEK Ondřej, BRUMMER Niko, DE Villiers Edward and LAFACE Pietro. Gender Independent Discriminative Speaker Recognition in I-Vector Space. In: Proc. International Conference on Acoustics, Speech, and Signal P. Kyoto: IEEE Signal Processing Society, 2012, pp. 4361-4364. ISBN 978-1-4673-0044-5.
Publication language:english
Original title:Gender Independent Discriminative Speaker Recognition in I-Vector Space
Title (cs):Diskriminativní rozpoznávání mluvčího v i-vektorovém prostoru nezávislé na pohlaví
Pages:4361-4364
Proceedings:Proc. International Conference on Acoustics, Speech, and Signal P
Conference:The 37th International Conference on Acoustics, Speech, and Signal Processing
Place:Kyoto, JP
Year:2012
ISBN:978-1-4673-0044-5
Publisher:IEEE Signal Processing Society
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2012/cumani_icassp2012_0004361.pdf [PDF]
Keywords
speaker recognition, gender recognition, PLDA models, GI Pairwise SVM
Annotation
This paper describes speaker recognition systems that are trained with gender dependent features and tested with known gender trails.
Abstract
Speaker recognition systems attain their best accuracy when trained with gender dependent features and tested with known gender trials. In real applications, howevcer, gender labels are often not given. In this work we illustrate the design of a system that does not make use of the gender labels both in training and in test, i.e. a completely Gender Independent (GI) system. It relies on discriminative training, where the trials are i-vector pairs, and the discrimination is between the hypothesis that the pair of feature vectors in the trial belong to the same speaker or to different speakers. We demonstrate that this pairwise discriminative training can be interpreted as a procedure that estimates the parameters of the best (second order) approximation of the log-likelihood ratio score function, and that a pairwise SVM can be used for training a gender independent system. Our results show that a pairwise GI SVM, saving memory and execution time, achieves on the last NIST evaluationscomplet state-of-the-art performance, comparable to a Gender Dependent(GD) system.
BibTeX:
@INPROCEEDINGS{
   author = {Sandro Cumani and Ond{\v{r}}ej Glembek and Niko Brummer and
	Edward Villiers de and Pietro Laface},
   title = {Gender Independent Discriminative Speaker Recognition in
	I-Vector Space},
   pages = {4361--4364},
   booktitle = {Proc. International Conference on Acoustics, Speech, and
	Signal P},
   year = {2012},
   location = {Kyoto, JP},
   publisher = {IEEE Signal Processing Society},
   ISBN = {978-1-4673-0044-5},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php.en.iso-8859-2?id=9942}
}

Your IPv4 address: 54.92.194.75
Switch to IPv6 connection

DNSSEC [dnssec]