Zobrazeno 1 - 10
of 66
pro vyhledávání: '"Roberto Gemello"'
Publikováno v:
INTERSPEECH
In this paper, we apply Semi-Supervised Learning (SSL) along with Data Augmentation (DA) for improving the accuracy of End-to-End ASR. We focus on the consistency regularization principle, which has been successfully applied to image classification t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4bc46f7beed87cdad035d8f2200424be
Publikováno v:
ASRU
Deep Neural Network (DNN) acoustic models are sensitive to the mismatch between training and testing environments. When a trained model is tested on unseen speakers, domain, or environment, recognition accuracy can degrade substantially. In such a ca
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1d61f8df45df282663c6293bd50c1df7
https://doi.org/10.1109/asru46091.2019.9003883
https://doi.org/10.1109/asru46091.2019.9003883
Publikováno v:
Pattern Recognition Letters. 31:1302-1309
In this paper we describe the implementation of a complete ANN training procedure using the block mode back-propagation learning algorithm for sequential patterns - such as the observation feature vectors of a speech recognition system - exploiting t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f61919b8d3160e15bdc18a8790e99263
https://doi.org/10.1016/j.patrec.2010.02.003
https://doi.org/10.1016/j.patrec.2010.02.003
Publikováno v:
Speech Communication. 49:827-835
This paper focuses on the adaptation of Automatic Speech Recognition systems using Hybrid models combining Artificial Neural Networks (ANN) with Hidden Markov Models (HMM). Most adaptation techniques for ANNs reported in the literature consist in add
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6770b5a23ab5c2f4717e1a5fd2325d6d
https://doi.org/10.1016/j.specom.2006.11.005
https://doi.org/10.1016/j.specom.2006.11.005
Publikováno v:
IEEE Signal Processing Letters. 13:56-59
A soft decision gain modification is introduced and applied to the Ephraim-Malah gain function based on maximum mean-square-error estimation after amplitude compression. Nonlinear evaluations of the noise overestimation factor and spectral floor are
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2d9a8349b9f81eadd38787a221bb71dd
https://doi.org/10.1109/lsp.2005.860535
https://doi.org/10.1109/lsp.2005.860535
Publikováno v:
Computer Speech & Language. 20:2-21
This paper investigates the potential of exploiting the redundancy implicit in multiple resolution analysis for automatic speech recognition systems. The analysis is performed by a binary tree of elements, each one of which is made by a half-band fil
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::de8b5da5804b160efb3bcdf1e6ab4001
https://doi.org/10.1016/j.csl.2004.06.001
https://doi.org/10.1016/j.csl.2004.06.001
Autor:
Hung H. Bui, Andrew P. Breen, Roberto Gemello, Peter F. Patel-Schneider, Richard S. Crouch, Adwait Ratnaparkhi, Charles L. Ortiz, Peter Stubley, Jiaying Shen, Paul van Mulbregt, Ronald M. Kaplan, William F. Ganong, Holger Quast, Kevin R. Farrell, Friedrich Faubel, Tim Haulick, Vlad Sejnoha
Publikováno v:
Interactive Displays: Natural Human-Interface Technologies
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::948deb3c4e8788e05db243da9ffba46a
https://doi.org/10.1002/9781118706237.ch3
https://doi.org/10.1002/9781118706237.ch3
Publikováno v:
Computer Speech & Language. 15:341-354
This paper describes an investigation on the possibility of adding new features to classical Mel Scaled Cepstral Coefficients (MFCC) and their time derivatives. A hybrid Automatic Speech Recognition (ASR) system is used based on a Neural Network (NN)
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::70b0c7c9c77f732f23c8b0143a08be3c
https://doi.org/10.1006/csla.2001.0171
https://doi.org/10.1006/csla.2001.0171
Publikováno v:
Information Sciences. 123:3-11
This paper describes the benefits in recognition accuracy that can be achieved in a hybrid Hidden Markov Model – Neural Network (HMM–NN) recognition framework by using context-dependent subword units named Stationary–Transitional Units. These u
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a948acd2cb8a5f6e0141db893bacf2de
https://doi.org/10.1016/s0020-0255(99)00106-1
https://doi.org/10.1016/s0020-0255(99)00106-1
Publikováno v:
Neural Processing Letters. 8:163-179
At present the Multi-layer Perceptron model (MLP) is, without doubt, the most diffused neural network for applications. So, it is important, from an engineering point of view, to design and test methods to improve MLP efficiency at run time. This pap
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cf73a3c3d26528b364344a976bbda7b7
https://doi.org/10.1023/a:1009600830022
https://doi.org/10.1023/a:1009600830022