Publications associées à Gradient-based spectral visualization of CNNs using raw waveforms

Robust Audio Segmentation

Audio segmentation, in general, is the task of segmenting a continuous audio stream in terms of acoustically homogenous regions, where the rule of homogeneity depends on the task. This thesis aims at developing and investigating efficient, robust and unsup ...

IDIAP2004

Robust Audio Segmentation

Hervé Bourlard, Jitendra Ajmera

Audio segmentation, in general, is the task of segmenting a continuous audio stream in terms of acoustically homogenous regions, where the rule of homogeneity depends on the task. This thesis aims at developing and investigating efficient, robust and unsup ...

École Polytechnique Fédérale de Lausanne2004

Spectro-Temporal Activity Pattern (STAP) Features for Noise Robust ASR

Hervé Bourlard, Hemant Misra, Shajith Ikbal

In this paper, we introduce a new noise robust representation of speech signal obtained by locating points of potential importance in the spectrogram, and parameterizing the activity of time-frequency pattern around those points. These features are referre ...

2004

Spectro-Temporal Activity Pattern (STAP) Features for Noise Robust ASR

Hervé Bourlard, Hemant Misra, Shajith Ikbal

In this paper, we introduce a new noise robust representation of speech signal obtained by locating points of potential importance in the spectrogram, and parameterizing the activity of time-frequency pattern around those points. These features are referre ...

IDIAP2004

Towards Robust and Adaptive Speech Recognition Models

Hervé Bourlard, Samy Bengio, Katrin Weber

In this paper, we discuss a family of new Automatic Speech Recognition (ASR) approaches, which somewhat deviate from the usual ASR approaches but which have recently been shown to be more robust to nonstationary noise, without requiring specific adaptation ...

Springer2004

Evaluation of Formant-Like Features for ASR

Hervé Bourlard, Samy Bengio, Katrin Weber

This paper investigates possibilities to automatically find a low-dimensional, formant-related physical representation of the speech signal, which is suitable for automatic speech recognition (ASR). This aim is motivated by the fact that formants have been ...

IDIAP2002

Evaluation of Formant-Like Features for ASR

Hervé Bourlard, Samy Bengio, Katrin Weber

This paper investigates possibilities to automatically find a low-dimensional, formant-related physical representation of the speech signal, which is suitable for automatic speech recognition (ASR). This aim is motivated by the fact that formants have been ...

2002

Analytic Assessment of Telephone Transmission Impact on ASR Performance Using a Simulation Model

Hervé Bourlard

This paper addresses the impact of telephone transmission channels on automatic speech recognition (ASR) performance. A real-time simulation model is described and implemented, which allows impairments that are encountered in traditional as well as modern ...

2002

TODE: A Decoder for Continuous Speech Recognition

Darren Moore

This document describes a new continuous speech decoder, TODE, which is compatible with the Torch machine learning software library. A brief theory of speech recognition is presented followed by a detailed description of the architecture of TODE and the co ...

IDIAP2002

HMM2- Extraction of Formant Features and their Use for Robust ASR

Hervé Bourlard, Samy Bengio, Katrin Weber

As recently introduced, an HMM2 can be considered as a particular case of an HMM mixture in which the HMM emission probabilities (usually estimated through Gaussian mixtures or an artificial neural network) are modeled by state-dependent, feature-based HMM ...

2001

Gradient-based spectral visualization of CNNs using raw waveforms

Graph Chatbot

Chattez avec Graph Search

Robust Audio Segmentation

Robust Audio Segmentation

Spectro-Temporal Activity Pattern (STAP) Features for Noise Robust ASR

Spectro-Temporal Activity Pattern (STAP) Features for Noise Robust ASR

Towards Robust and Adaptive Speech Recognition Models

Evaluation of Formant-Like Features for ASR

Evaluation of Formant-Like Features for ASR

Analytic Assessment of Telephone Transmission Impact on ASR Performance Using a Simulation Model

TODE: A Decoder for Continuous Speech Recognition

HMM2- Extraction of Formant Features and their Use for Robust ASR

Robust Audio Segmentation

Spectro-Temporal Activity Pattern (STAP) Features for Noise Robust ASR

Robust Audio Segmentation

Towards Robust and Adaptive Speech Recognition Models

Spectro-Temporal Activity Pattern (STAP) Features for Noise Robust ASR

Evaluation of Formant-Like Features for ASR

HMM2- Extraction of Formant Features and their Use for Robust ASR

Analytic Assessment of Telephone Transmission Impact on ASR Performance Using a Simulation Model

TODE: A Decoder for Continuous Speech Recognition

Evaluation of Formant-Like Features for ASR