Článek ve sborníku konference

SILNOVA Anna, BURGET Lukáš a ČERNOCKÝ Jan. Alternative Approaches to Neural Network based Speaker Verification. In: Proceedings of Interspeech 2017. Stockholm: International Speech Communication Association, 2017, s. 1572-1575. ISSN 1990-9772. Dostupné z: http://www.isca-speech.org/archive/Interspeech_2017/pdfs/1062.PDF
Jazyk publikace:angličtina
Název publikace:Alternative Approaches to Neural Network based Speaker Verification
Název (cs):Alternativní přístupy k neuronovým sítím založené na ověřování řečníka
Strany:1572-1575
Sborník:Proceedings of Interspeech 2017
Konference:Interspeech 2017
Místo vydání:Stockholm, SE
Rok:2017
URL:http://www.isca-speech.org/archive/Interspeech_2017/pdfs/1062.PDF
Časopis:Proceedings of Interspeech, roč. 2017, č. 08, FR
ISSN:1990-9772
DOI:10.21437/Interspeech.2017-1062
Vydavatel:International Speech Communication Association
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2017/silnova_interspeech2017_IS171062.pdf [PDF]
Klíčová slova
automatic speaker recognition, deep neural networks, bottleneck features
Anotace
Tento článek pojednává o alternativních přístupech k neuronovým sítím, kzeré je založené na ověřování řečníka.
Abstrakt
Just like in other areas of automatic speech processing, feature extraction based on bottleneck neural networks was recently found very effective for the speaker verification task. However, better results are usually reported with more complex neural network architectures (e.g. stacked bottlenecks), which are difficult to reproduce. In this work, we experiment with the so called deep features, which are based on a simple feed-forward neural network architecture. We study various forms of applying deep features to i-vector/PDA based speaker verification. With proper settings, better verification performance can be obtained by means of this simple architecture as compared to the more elaborate bottleneck features. Also, we further experiment with multi-task training, where the neural network is trained for both speaker recognition and senone recognition objectives. Results indicate that, with a careful weighting of the two objectives, multi-task training can result in significantly better performing deep features.
BibTeX:
@INPROCEEDINGS{
   author = {Anna Silnova and Luk{\'{a}}{\v{s}} Burget and Jan
	{\v{C}}ernock{\'{y}}},
   title = {Alternative Approaches to Neural Network based
	Speaker Verification},
   pages = {1572--1575},
   booktitle = {Proceedings of Interspeech 2017},
   journal = {Proceedings of Interspeech},
   volume = {2017},
   number = {08},
   year = {2017},
   location = {Stockholm, SE},
   publisher = {International Speech Communication Association},
   ISSN = {1990-9772},
   doi = {10.21437/Interspeech.2017-1062},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php.cs?id=11582}
}

Vaše IPv4 adresa: 3.80.223.123