Performances of an automatic speech recognition system degrade when test and training conditions do not match. Classical Stochastic Matching (SM) method proposes an off-line estimation of a compensation function that maximizes the likelihood of the compensated speech, given the optimal sequence of models proposed by the recognition process. We developed a new frame-synchronous technic based on SM : compensation is performed in parallel with the recognition. This is suitable to cope with slowly varying noise. We proposed two additional versions of our approach: -a tree structure of transformations is used to build a state-dependant non-linear compensation function. This is motivated by the fact that similar observations will be affected simi...
Colloque avec actes et comité de lecture. nationale.National audienceWe present a new predictive com...
We present a sequential Monte Carlo method applied to additive noise compensation for robust speech ...
We present a sequential Monte Carlo method applied to additive noise compensation for robust speech ...
Performances of an automatic speech recognition system degrade when test and training conditions do ...
Les performances d'un système de reconnaissance automatique de la parole se dégradent lorsque les co...
Colloque avec actes sans comité de lecture. internationale.International audienceIn real life speech...
Most automatic speech recognisers perform poorly when the training and testing conditions are not ma...
Colloque avec actes et comité de lecture. nationale.National audienceAn improvement of a previously ...
[[abstract]]© 2001 Institute of Electrical and Electronics Engineers - In this paper, an signal-to-n...
Colloque avec actes et comité de lecture. internationale.International audienceIn this paper we pres...
While both the acoustic model and the language model in automatic speech recognition are typically w...
Dans cette thèse nous élaborons quatre composantes fondamentales d'un système de reconnaissance auto...
The role of a stochastic language model is to give the best estimation possible of the probability o...
Colloque avec actes et comité de lecture. internationale.International audienceWe present a frame-sy...
In this paper we investigate how to improve the robustness of a speech recognizer in a noisy, mismat...
Colloque avec actes et comité de lecture. nationale.National audienceWe present a new predictive com...
We present a sequential Monte Carlo method applied to additive noise compensation for robust speech ...
We present a sequential Monte Carlo method applied to additive noise compensation for robust speech ...
Performances of an automatic speech recognition system degrade when test and training conditions do ...
Les performances d'un système de reconnaissance automatique de la parole se dégradent lorsque les co...
Colloque avec actes sans comité de lecture. internationale.International audienceIn real life speech...
Most automatic speech recognisers perform poorly when the training and testing conditions are not ma...
Colloque avec actes et comité de lecture. nationale.National audienceAn improvement of a previously ...
[[abstract]]© 2001 Institute of Electrical and Electronics Engineers - In this paper, an signal-to-n...
Colloque avec actes et comité de lecture. internationale.International audienceIn this paper we pres...
While both the acoustic model and the language model in automatic speech recognition are typically w...
Dans cette thèse nous élaborons quatre composantes fondamentales d'un système de reconnaissance auto...
The role of a stochastic language model is to give the best estimation possible of the probability o...
Colloque avec actes et comité de lecture. internationale.International audienceWe present a frame-sy...
In this paper we investigate how to improve the robustness of a speech recognizer in a noisy, mismat...
Colloque avec actes et comité de lecture. nationale.National audienceWe present a new predictive com...
We present a sequential Monte Carlo method applied to additive noise compensation for robust speech ...
We present a sequential Monte Carlo method applied to additive noise compensation for robust speech ...