This thesis work aims to respond to a request initiated by the company Ixiade. The request was to explore Natural language processing methods in order to propose a content classification tool. Two types of data were used throughout the study : interview transcripts and online data. Both came from studies carries out to assess the acceptability of an innovation.This research work uses data amplification methods combined with Transformer-based-models to classify transcribed oral data and online data stemming from a community platform. The contributions are as follows: (1) Proposal of a methodology to build a training corpus in a context where data are unavailable; (2) Proposal of a method for extracting and filtering textual content accordin...
A study on disfluencies in oral French utterances has been undertaken for 5 years. The oral data tra...
Les technologies liées à la parole, et en particulier la reconnaissance de la parole, suscitent un g...
Our researches are based upon the EPAC project. We develop this work context in our first chapter. T...
This thesis is part of a study that explores automatic transcription potential for the instrumentati...
Application of spoken language understanding aim to extract relevant items of meaning from spoken si...
Two wide research fields named Speech Recognition and Machine Learning meet with the Automatic Speec...
Corpora, which are text collections selected for specific purposes, are playing an increasing role i...
This thesis tackles the problem of processing data derived from the oral. Indeed, businesses are ful...
While huge progress has been made in machine translation (MT) in recent years, the majority of MT sy...
Cette thèse s’inscrit dans le cadre d’une étude sur le potentiel de la transcription automatique pou...
In statistical language modelling researches, there is a lack of huge text corpora, especially for s...
This thesis is a part of the emergence of deep learning and focuses on spoken language understanding...
In a world where a considerable number of complex systems and smart objects are emerging, the need t...
Colloque avec actes et comité de lecture. internationale.International audienceThis paper presents a...
Les applications de Traitement Automatique des Langues nécessitent le plus souvent des corpus homogè...
A study on disfluencies in oral French utterances has been undertaken for 5 years. The oral data tra...
Les technologies liées à la parole, et en particulier la reconnaissance de la parole, suscitent un g...
Our researches are based upon the EPAC project. We develop this work context in our first chapter. T...
This thesis is part of a study that explores automatic transcription potential for the instrumentati...
Application of spoken language understanding aim to extract relevant items of meaning from spoken si...
Two wide research fields named Speech Recognition and Machine Learning meet with the Automatic Speec...
Corpora, which are text collections selected for specific purposes, are playing an increasing role i...
This thesis tackles the problem of processing data derived from the oral. Indeed, businesses are ful...
While huge progress has been made in machine translation (MT) in recent years, the majority of MT sy...
Cette thèse s’inscrit dans le cadre d’une étude sur le potentiel de la transcription automatique pou...
In statistical language modelling researches, there is a lack of huge text corpora, especially for s...
This thesis is a part of the emergence of deep learning and focuses on spoken language understanding...
In a world where a considerable number of complex systems and smart objects are emerging, the need t...
Colloque avec actes et comité de lecture. internationale.International audienceThis paper presents a...
Les applications de Traitement Automatique des Langues nécessitent le plus souvent des corpus homogè...
A study on disfluencies in oral French utterances has been undertaken for 5 years. The oral data tra...
Les technologies liées à la parole, et en particulier la reconnaissance de la parole, suscitent un g...
Our researches are based upon the EPAC project. We develop this work context in our first chapter. T...