Zobrazeno 1 - 10
of 60
pro vyhledávání: '"Silovsky, Jan"'
Autor:
Pelikan, Martin, Azam, Sheikh Shams, Feldman, Vitaly, Silovsky, Jan "Honza", Talwar, Kunal, Likhomanenko, Tatiana
While federated learning (FL) has recently emerged as a promising approach to train machine learning models, it is limited to only preliminary explorations in the domain of automatic speech recognition (ASR). Moreover, FL does not inherently guarante
Externí odkaz:
http://arxiv.org/abs/2310.00098
In this paper, we start by training End-to-End Automatic Speech Recognition (ASR) models using Federated Learning (FL) and examining the fundamental considerations that can be pivotal in minimizing the performance gap in terms of word error rate betw
Externí odkaz:
http://arxiv.org/abs/2309.13102
Autor:
Silovsky, Jan, Deng, Liuhui, Argueta, Arturo, Arvizo, Tresi, Hsiao, Roger, Kuznietsov, Sasha, Lin, Yiu-Chang, Xiao, Xiaoqiang, Zhang, Yuanyuan
Voice technology has become ubiquitous recently. However, the accuracy, and hence experience, in different languages varies significantly, which makes the technology not equally inclusive. The availability of data for different languages is one of th
Externí odkaz:
http://arxiv.org/abs/2305.13652
Autor:
Jiang, Zhuolin, Silovsky, Jan, Siu, Man-Hung, Hartmann, William, Gish, Herbert, Adali, Sancar
Multi-label image classification has generated significant interest in recent years and the performance of such systems often suffers from the not so infrequent occurrence of incorrect or missing labels in the training data. In this paper, we extend
Externí odkaz:
http://arxiv.org/abs/2005.00596
Spoken language identification (LID) technologies have improved in recent years from discriminating largely distinct languages to discriminating highly similar languages or even dialects of the same language. One aspect that has been mostly neglected
Externí odkaz:
http://arxiv.org/abs/2001.11019
Autor:
Gish, Herbert, Silovsky, Jan, Sung, Man-Ling, Siu, Man-Hung, Hartmann, William, Jiang, Zhuolin
We investigate the problem of machine learning with mislabeled training data. We try to make the effects of mislabeled training better understood through analysis of the basic model and equations that characterize the problem. This includes results a
Externí odkaz:
http://arxiv.org/abs/1909.09136
Publikováno v:
Transitions Online. 3/6/2017, p1-1. 1p.
Publikováno v:
In Speech Communication November-December 2013 55(10):1033-1046
Enhancement of emotion detection in spoken dialogue systems by combining several information sources
Publikováno v:
In Speech Communication 2011 53(9):1210-1228
Autor:
Nouza, Jan, Silovsky, Jan
Publikováno v:
Radioengineering, Vol 18, Iss 4, Pp 665-670 (2009)
Radioengineering. 2009, vol. 18, č. 4, s. 665-670. ISSN 1210-2512
Radioengineering. 2009, vol. 18, č. 4, s. 665-670. ISSN 1210-2512
In the paper, we present a system designed for detecting keywords in telephone speech. We focus not only on achieving high accuracy but also on very short processing time. The keyword spotting system can run in three modes: a) an off-line mode requir