Zobrazeno 1 - 10
of 289
pro vyhledávání: '"Nóvoa, José"'
Autor:
Novoa, José, Mahu, Rodrigo, Díaz, Alejandro, Wuth, Jorge, Stern, Richard, Yoma, Nestor Becerra
This paper describes the integration of weighted delay-and-sum beamforming with speech source localization using image processing and robot head visual servoing for source tracking. We take into consideration the fact that the directivity gain provid
Externí odkaz:
http://arxiv.org/abs/1906.07298
Autor:
Novoa, José, Fredes, Josué, Wuth, Jorge, Huenupán, Fernando, Stern, Richard M., Yoma, Nestor Becerra
This paper addresses the combination of complementary parallel speech recognition systems to reduce the error rate of speech recognition systems operating in real highly-reverberant environments. First, the testing environment consists of recordings
Externí odkaz:
http://arxiv.org/abs/1906.07299
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Autor:
Novoa, José, Escudero, Juan Pablo, Wuth, Jorge, Poblete, Victor, King, Simon, Stern, Richard, Yoma, Néstor Becerra
This paper evaluates the robustness of a DNN-HMM-based speech recognition system in highly-reverberant real environments using the HRRE database. The performance of locally-normalized filter bank (LNFB) and Mel filter bank (MelFB) features in combina
Externí odkaz:
http://arxiv.org/abs/1803.09013
Autor:
Escudero, Juan Pablo, Novoa, José, Mahu, Rodrigo, Wuth, Jorge, Huenupán, Fernando, Stern, Richard, Yoma, Néstor Becerra
Reverberation and additive noise have detrimental effects on the performance of automatic speech recognition systems. In this paper we explore the ability of a DNN-based spectral feature mapping to remove the effects of reverberation and additive noi
Externí odkaz:
http://arxiv.org/abs/1803.09016
Autor:
Escudero, Juan Pablo, Poblete, Victor, Novoa, José, Wuth, Jorge, Fredes, Josué, Mahu, Rodrigo, Stern, Richard, Yoma, Néstor Becerra
Speech recognition in highly-reverberant real environments remains a major challenge. An evaluation dataset for this task is needed. This report describes the generation of the Highly-Reverberant Real Environment database (HRRE). This database contai
Externí odkaz:
http://arxiv.org/abs/1801.09651
Autor:
Novoa, José, Escudero, Juan Pablo, Fredes, Josué, Wuth, Jorge, Mahu, Rodrigo, Yoma, Néstor Becerra
In real human robot interaction (HRI) scenarios, speech recognition represents a major challenge due to robot noise, background noise and time-varying acoustic channel. This document describes the procedure used to obtain the Multichannel Robot Speec
Externí odkaz:
http://arxiv.org/abs/1801.00061
In this paper, the uncertainty is defined as the mean square error between a given enhanced noisy observation vector and the corresponding clean one. Then, a DNN is trained by using enhanced noisy observation vectors as input and the uncertainty as o
Externí odkaz:
http://arxiv.org/abs/1705.10368
Autor:
Martínez, Carlos, Teijeira, Susana, Domínguez, Patricia, Campanioni, Silvia, Busto, Laura, González-Nóvoa, José A., Alonso, Jacobo, Poveda, Eva, San Millán, Beatriz, Veiga, César
Publikováno v:
Machine Learning & Knowledge Extraction; Sep2024, Vol. 6 Issue 3, p2018-2032, 15p
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.