Zobrazeno 1 - 10
of 36
pro vyhledávání: '"Vitaladevuni, Shiv"'
Autor:
Zeng, Lu, Parthasarathi, Sree Hari Krishnan, Liu, Yuzong, Escott, Alex, Cheekatmalla, Santosh Kumar, Strom, Nikko, Vitaladevuni, Shiv
We propose a novel 2-stage sub 8-bit quantization aware training algorithm for all components of a 250K parameter feedforward, streaming, state-free keyword spotting model. For the 1st-stage, we adapt a recently proposed quantization technique using
Externí odkaz:
http://arxiv.org/abs/2207.06920
Autor:
Gao, Yixin, Stein, Noah D., Kao, Chieh-Chi, Cai, Yunliang, Sun, Ming, Zhang, Tao, Vitaladevuni, Shiv
Wake word (WW) spotting is challenging in far-field due to the complexities and variations in acoustic conditions and the environmental interference in signal transmission. A suite of carefully designed and optimized audio front-end (AFE) algorithms
Externí odkaz:
http://arxiv.org/abs/2010.06676
Publikováno v:
Proc. ICASSP 2020
Wake word (WW) spotting is challenging in far-field not only because of the interference in signal transmission but also the complexity in acoustic environments. Traditional WW model training requires large amount of in-domain WW-specific data with s
Externí odkaz:
http://arxiv.org/abs/2010.06659
Autor:
Jose, Christin, Mishchenko, Yuriy, Senechal, Thibaud, Shah, Anish, Escott, Alex, Vitaladevuni, Shiv
Publikováno v:
Interspeech 2020
Small footprint embedded devices require keyword spotters (KWS) with small model size and detection latency for enabling voice assistants. Such a keyword is often referred to as \textit{wake word} as it is used to wake up voice assistant enabled devi
Externí odkaz:
http://arxiv.org/abs/2008.03790
This paper proposes a Sub-band Convolutional Neural Network for spoken term classification. Convolutional neural networks (CNNs) have proven to be very effective in acoustic applications such as spoken term classification, keyword spotting, speaker i
Externí odkaz:
http://arxiv.org/abs/1907.01448
Autor:
Sun, Ming, Raju, Anirudh, Tucker, George, Panchapagesan, Sankaran, Fu, Gengshen, Mandal, Arindam, Matsoukas, Spyros, Strom, Nikko, Vitaladevuni, Shiv
Publikováno v:
Spoken Language Technology Workshop (SLT), 2016 IEEE (pp. 474-480). IEEE
We propose a max-pooling based loss function for training Long Short-Term Memory (LSTM) networks for small-footprint keyword spotting (KWS), with low CPU, memory, and latency requirements. The max-pooling loss training can be further guided by initia
Externí odkaz:
http://arxiv.org/abs/1705.02411
Thesis (Ph. D.) -- University of Maryland, College Park, 2007.
Thesis research directed by: Dept. of Computer Science. Title from t.p. of PDF. Includes bibliographical references. Published by UMI Dissertation Services, Ann Arbor, Mich. Also ava
Thesis research directed by: Dept. of Computer Science. Title from t.p. of PDF. Includes bibliographical references. Published by UMI Dissertation Services, Ann Arbor, Mich. Also ava
Externí odkaz:
http://hdl.handle.net/1903/7610
Autor:
Hu, Tao, Nunez-Iglesias, Juan, Vitaladevuni, Shiv, Scheffer, Lou, Xu, Shan, Bolorizadeh, Mehdi, Hess, Harald, Fetter, Richard, Chklovskii, Dmitri
A central problem in neuroscience is reconstructing neuronal circuits on the synapse level. Due to a wide range of scales in brain architecture such reconstruction requires imaging that is both high-resolution and high-throughput. Existing electron m
Externí odkaz:
http://arxiv.org/abs/1210.0564
Publikováno v:
2015 IEEE 14th International Conference on Machine Learning & Applications (ICMLA); 2015, p369-374, 6p
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.