Publication

Robust speech recognition based on multi-stream processing

Publications associées (74)

Graph Chatbot

Chattez avec Graph Search

Posez n’importe quelle question sur les cours, conférences, exercices, recherches, actualités, etc. de l’EPFL ou essayez les exemples de questions ci-dessous.

AVERTISSEMENT : Le chatbot Graph n'est pas programmé pour fournir des réponses explicites ou catégoriques à vos questions. Il transforme plutôt vos questions en demandes API qui sont distribuées aux différents services informatiques officiellement administrés par l'EPFL. Son but est uniquement de collecter et de recommander des références pertinentes à des contenus que vous pouvez explorer pour vous aider à répondre à vos questions.

Connectez-vous pour utiliser Chat avec Graph Search

SURE-LET interscale-intercolor wavelet thresholding for color image denoising - art. no. 67011H

Thierry Blu, Florian Luisier

We propose a new orthonormal wavelet thresholding algorithm for denoising color images that are assumed to be corrupted by additive Gaussian white noise of known intercolor covariance matrix. The proposed wavelet denoiser consists of a linear expansion of ...

Spie-Int Soc Optical Engineering, Po Box 10, Bellingham, Wa 98227-0010 Usa2007

Noise Modeling in Lateral Nonuniform MOSFET

Christian Enz, Jean-Michel Sallese, Ananda Sankar Roy

In this paper, we present an analytical noise modeling methodology for lateral nonuniform MOSFET. We demonstrate that the noise properties of lateral nonuniform MOSFETs are considerably different from the prediction obtained with the conventional Klaassen- ...

2007

Noise and small-signal modeling of nanoscale MOSFETs

Ananda Sankar Roy

After years of intensive research effort, the design of RF integrated circuits in CMOS has now reached a wide acceptance for industrial designs. This is due to the high unity gain frequency and low-noise performance of today's deep sub micrometer MOS trans ...

EPFL2007

Unsupervised Spectral Subtraction for Noise-Robust ASR on Unknown Transmission Channels

Hervé Bourlard, Guillaume Lathoud

This paper addresses several issues of classical spectral subtraction methods with respect to the automatic speech recognition task in noisy environments. The main contributions of this paper are twofold. First, a channel normalization method is proposed t ...

IDIAP2006

Harmonic Plus Noise Model for Concatenative Speech Synthesis

This project develops the new model Harmonic Plus Noise applied for the concatenative speech synthesis. The software is composed of an analysis part (off-line process) applied on the first initial database and a synthesis part (real time process) applied o ...

IDIAP2005

On the quadratic AWGN CEO problem and non-Gaussian sources

Michael Christoph Gastpar

In the CEO problem, introduced by Berger et al, IEEE, Trans. Info. Theory, 1996, a CEO is interested in a source that cannot be observed directly. M agents observe independently noisy versions of the source and, without collaborating, must encode these acr ...

Ieee Service Center, 445 Hoes Lane, Po Box 1331, Piscataway, Nj 08855-1331 Usa2005

Generalized Smoothing Splines and the Optimal Discretization of the Wiener Filter

Michaël Unser, Thierry Blu

We introduce an extended class of cardinal LL-splines, where L is a pseudo-differential operator satisfying some admissibility conditions. We show that the LL-spline signal interpolation problem is well posed and that its solution is the unique minimizer ...

IEEE2005

Unsupervised Spectral Substraction for Noise-Robust ASR

Hervé Bourlard, Guillaume Lathoud, Bertrand Mesot

This paper proposes a simple, computationally efficient \mbox{2-mixture} model approach to discriminate between speech and background noise at the magnitude spectrogram level. It is directly derived from observations on real data, and can be used in a full ...

IDIAP2005

Unsupervised Spectral Subtraction for Noise-Robust ASR

Hervé Bourlard, Guillaume Lathoud, Bertrand Mesot

This paper proposes a simple, computationally efficient 2-mixture model approach to discriminate between speech and background noise at the magnitude spectrogram level. It is directly derived from observations on real data, and can be used in a fully unsup ...

2005

A Frequency-Domain Silence Noise Model

Guillaume Lathoud, Bertrand Mesot

This paper proposes a simple, computationally efficient 2-mixture model approach to discrimination between speech and background noise. It is directly derived from observations on real data, and can be used in a fully unsupervised manner, with the EM algor ...

IDIAP2005