Zobrazeno 1 - 10
of 84
pro vyhledávání: '"Vaibhava Goel"'
Publikováno v:
Findings of the Association for Computational Linguistics: EMNLP 2021.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::749cf6e0b599b4cd19db6270339d74cd
https://doi.org/10.18653/v1/2021.findings-emnlp.87
https://doi.org/10.18653/v1/2021.findings-emnlp.87
Publikováno v:
EMNLP (1)
BERT-era question answering systems have recently achieved impressive performance on several question-answering (QA) tasks. These systems are based on representations that have been pre-trained on self-supervised tasks such as word masking and senten
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a549899661655e4d92c3c86270483ed4
https://doi.org/10.18653/v1/2020.emnlp-main.87
https://doi.org/10.18653/v1/2020.emnlp-main.87
Autor:
Vaibhava Goel, Xiaodong Cui
Publikováno v:
ICASSP
Feature transformations are commonly used in speech recognition to account for distribution mismatches between the source and target domains also referred to as covariate shift. Linear affine or piecewise linear transformations are typically consider
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8e17bfd60c03efa5a2b49dba9c4be1ea
https://doi.org/10.1109/taslp.2016.2594255
https://doi.org/10.1109/taslp.2016.2594255
Publikováno v:
INTERSPEECH
An embedding-based speaker adaptive training (SAT) approach is proposed and investigated in this paper for deep neural network acoustic modeling. In this approach, speaker embedding vectors, which are a constant given a particular speaker, are mapped
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fccb095c63115e7d1811ce3c8f1574d5
http://arxiv.org/abs/1710.06937
http://arxiv.org/abs/1710.06937
Publikováno v:
CVPR
Recently it has been shown that policy-gradient methods for reinforcement learning can be utilized to train deep end-to-end systems directly on non-differentiable metrics for the task at hand. In this paper we consider the problem of optimizing image
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::92178445c3edf1d61d7b4974c648338a
https://doi.org/10.1109/cvpr.2017.131
https://doi.org/10.1109/cvpr.2017.131
Autor:
Spyridon Thermos, Gerasimos Potamianos, Alexandros Koumbaroulis, Vaibhava Goel, Etienne Marcheret, Youssef Mroueh, Argyrios Vartholomaios
Publikováno v:
The Handbook of Multimodal-Multisensor Interfaces, Volume 1 (1)
Chances are that most of us have experienced difficulty in listening to our interlocutor during face-to-face conversation while in highly noisy environments, such as next to heavy traffic or over the background of high-intensity speech babble or loud
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::aba79f96e71b496f941bce00d5fc3457
https://doi.org/10.1145/3015783.3015797
https://doi.org/10.1145/3015783.3015797
Publikováno v:
ICASSP
Dropout, the random dropping out of activations according to a specified rate, is a very simple but effective method to avoid over-fitting of deep neural networks to the training data.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f4f35191d358a4ee167fd79bdd3b7be0
https://doi.org/10.1109/icassp.2017.7952572
https://doi.org/10.1109/icassp.2017.7952572
Publikováno v:
Scopus-Elsevier
We introduce new families of Integral Probability Metrics (IPM) for training Generative Adversarial Networks (GAN). Our IPMs are based on matching statistics of distributions embedded in a finite dimensional feature space. Mean and covariance feature
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b232c3e10777bdf10f565995ef3d08b5
Publikováno v:
IEEE Transactions on Audio, Speech, and Language Processing. 21:2255-2266
We introduce a new class of parameter estimation methods for log-linear models. Our approach relies on the fact that minimizing a rational function of mixtures of exponentials is equivalent to minimizing a difference of convex functions. This allows
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::80507ec0459521ddaddf9de2295f457e
https://doi.org/10.1109/tasl.2013.2271592
https://doi.org/10.1109/tasl.2013.2271592
Publikováno v:
INTERSPEECH
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::106c3e0f75be05b3abaf8332df915a89
https://doi.org/10.21437/interspeech.2016-212
https://doi.org/10.21437/interspeech.2016-212