Conference paper

ROHDIN Johan A., SILNOVA Anna, DIEZ Sánchez Mireia, PLCHOT Oldřich, MATĚJKA Pavel and BURGET Lukáš. End-to-End DNN Based Speaker Recognition Inspired by i-Vector and PLDA. In: Proceedings of ICASSP. Calgary: IEEE Signal Processing Society, 2018, pp. 4874-4878. ISBN 978-1-5386-4658-8.
Publication language:english
Original title:End-to-End DNN Based Speaker Recognition Inspired by i-Vector and PLDA
Title (cs):End-to-end DNN rozpoznávání mluvčího inspirované i-vektory a PLDA
Pages:4874-4878
Proceedings:Proceedings of ICASSP
Conference:2018 IEEE International Conference on Acoustics, Speech and Signal Processing
Place:Calgary, CA
Year:2018
ISBN:978-1-5386-4658-8
Publisher:IEEE Signal Processing Society
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2018/rohdin_icassp2018_0004874.pdf [PDF]
Keywords
Speaker verification, DNN, end-to-end
Annotation
Recently, several end-to-end speaker verification systems based on deep neural networks (DNNs) have been proposed. These systems have been proven to be competitive for text-dependent tasks as well as for text-independent tasks with short utterances. However, for text-independent tasks with longer utterances, end-to-end systems are still outperformed by standard i-vector + PLDA systems. In this work, we develop an end-to-end speaker verification system that is initialized to mimic an i-vector + PLDA baseline. The system is then further trained in an end-to-end manner but regularized so that it does not deviate too far from the initial system. In this way we mitigate overfitting which normally limits the performance of endto- end systems. The proposed system outperforms the i-vector + PLDA baseline on both long and short duration utterances.
BibTeX:
@INPROCEEDINGS{
   author = {A. Johan Rohdin and Anna Silnova and Mireia
	S{\'{a}}nchez Diez and Old{\v{r}}ich Plchot and
	Pavel Mat{\v{e}}jka and Luk{\'{a}}{\v{s}} Burget},
   title = {End-to-End DNN Based Speaker Recognition Inspired
	by i-Vector and PLDA},
   pages = {4874--4878},
   booktitle = {Proceedings of ICASSP},
   year = {2018},
   location = {Calgary, CA},
   publisher = {IEEE Signal Processing Society},
   ISBN = {978-1-5386-4658-8},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=11724}
}

Your IPv4 address: 54.221.9.6
Switch to IPv6 connection

DNSSEC [dnssec]