National audienceThe topic of this study is automatic speech recognition and concerns more precisely speaker independent acoustic-phonetic decoding. The model that we present is roughly based on the human cognitive processing system: the speech signal excites several parallel distributed processes of decoding. All these processes are conceptually different and they consequently do not give the same output. The final decision is taken by an "intelligent" control process which studies all these responses. The global recognition process, which is a metric method, extracts acoutsic features using a P.L.P. algorithm (Perceptually based Linear Prediction). Then, the Data Time Warping method (DTW) allows us to spectrally compare stimuli to refere...
Recent advances in the availability of computational resources allow for more sophisticated approach...
We study in this thesis the joint construction of speech recognition and synthesis systems for new l...
La Reconnaissance Automatique de la Parole (RAP) est affectée par les nombreuses variabilités présen...
The aim of Speech Recognition is to identify with machines what a speaker is saying. This process ca...
Texte intégral accessible uniquement aux membres de l'Université de LorraineThe framework of this th...
L'une des difficultés d'une langue peu dotée est l'inexistence des services liés aux technologies du...
One of the difficulties of an unresourced language is the lack of technology services in the speech ...
This research study aims to present a retrospective study about speech recognition systems and artif...
International audienceWe aim at examining to what extent a knowledge-based model can recognise segme...
The discovery of elementary linguistic units (phonemes, words) only from sound recordings is an unre...
We propose a Multigranular Automatic Speech Recognizer. The hypothesis is that speech signal contai...
Automatic Speech Recognition (ASR) is affected by many variabilities present in the speech signal. D...
La découverte d'unités linguistiques élémentaires (phonèmes, mots) uniquement à partir d'enregistrem...
Les trois sources principalement utilisées en reconnaissance vocale automatique (Automatic Speech Re...
A quick assessment of the current state-of-the-art in Speech Recognition and Synthesis allows to not...
Recent advances in the availability of computational resources allow for more sophisticated approach...
We study in this thesis the joint construction of speech recognition and synthesis systems for new l...
La Reconnaissance Automatique de la Parole (RAP) est affectée par les nombreuses variabilités présen...
The aim of Speech Recognition is to identify with machines what a speaker is saying. This process ca...
Texte intégral accessible uniquement aux membres de l'Université de LorraineThe framework of this th...
L'une des difficultés d'une langue peu dotée est l'inexistence des services liés aux technologies du...
One of the difficulties of an unresourced language is the lack of technology services in the speech ...
This research study aims to present a retrospective study about speech recognition systems and artif...
International audienceWe aim at examining to what extent a knowledge-based model can recognise segme...
The discovery of elementary linguistic units (phonemes, words) only from sound recordings is an unre...
We propose a Multigranular Automatic Speech Recognizer. The hypothesis is that speech signal contai...
Automatic Speech Recognition (ASR) is affected by many variabilities present in the speech signal. D...
La découverte d'unités linguistiques élémentaires (phonèmes, mots) uniquement à partir d'enregistrem...
Les trois sources principalement utilisées en reconnaissance vocale automatique (Automatic Speech Re...
A quick assessment of the current state-of-the-art in Speech Recognition and Synthesis allows to not...
Recent advances in the availability of computational resources allow for more sophisticated approach...
We study in this thesis the joint construction of speech recognition and synthesis systems for new l...
La Reconnaissance Automatique de la Parole (RAP) est affectée par les nombreuses variabilités présen...