We present a system for audiovisual document structuring, based-on speaker role recognition and speech interaction zone detection. The first stage of our system consists in an automatic method for speech interaction zones detection and characterization. Such zones correspond to temporal sequences of documents which potentially contain conversations between speakers. The second stage of our system achieves the recognition of speaker roles : anchorman, journalist and other. Our contribution to this domain is based on the hypothesis that cues about speaker roles are available through low-level features extracted from the temporal organization of turn-takings and from acoustic and prosodic features (speech rate and pitch). In the last stage of o...
International audienceContrariwise to controlled speech, for which speaker's intention are very limi...
The aim of this work is to find prosodic markers of the hierarchical structure of a speech message f...
Le traitement automatique de la parole est un domaine qui englobe un grand nombre de travaux : de la...
Archives professionals have high expectations for efficient indexing tools. In particular, the purpo...
La tâche de Segmentation et Regroupement en Locuteurs (SRL), telle que définie par le NIST, considèr...
The increasing quantity of video material available requires the implementation of automatic structu...
La structuration en thèmes est un domaine de recherche très prisé dans le traitement automatique du ...
This thesis presents a work focusing on the topic of speaker diarization for different types of audi...
International audienceTo synthesize audiobooks in an expressive manner, it is necessary to know the ...
National audienceThe understanding of language mechanisms needs to take into account very precisely ...
This paper proposes an approach for the automatic recognition of roles in settings like news and tal...
La tâche de segmentation et de regroupement en locuteur (SRL) consiste à déterminer le nombre de loc...
This paper proposes an approach for the automatic recognition of roles in settings like news and tal...
Application of spoken language understanding aim to extract relevant items of meaning from spoken si...
New challenges emerged in the past years as the audiovisual landscape significantly transformed with...
International audienceContrariwise to controlled speech, for which speaker's intention are very limi...
The aim of this work is to find prosodic markers of the hierarchical structure of a speech message f...
Le traitement automatique de la parole est un domaine qui englobe un grand nombre de travaux : de la...
Archives professionals have high expectations for efficient indexing tools. In particular, the purpo...
La tâche de Segmentation et Regroupement en Locuteurs (SRL), telle que définie par le NIST, considèr...
The increasing quantity of video material available requires the implementation of automatic structu...
La structuration en thèmes est un domaine de recherche très prisé dans le traitement automatique du ...
This thesis presents a work focusing on the topic of speaker diarization for different types of audi...
International audienceTo synthesize audiobooks in an expressive manner, it is necessary to know the ...
National audienceThe understanding of language mechanisms needs to take into account very precisely ...
This paper proposes an approach for the automatic recognition of roles in settings like news and tal...
La tâche de segmentation et de regroupement en locuteur (SRL) consiste à déterminer le nombre de loc...
This paper proposes an approach for the automatic recognition of roles in settings like news and tal...
Application of spoken language understanding aim to extract relevant items of meaning from spoken si...
New challenges emerged in the past years as the audiovisual landscape significantly transformed with...
International audienceContrariwise to controlled speech, for which speaker's intention are very limi...
The aim of this work is to find prosodic markers of the hierarchical structure of a speech message f...
Le traitement automatique de la parole est un domaine qui englobe un grand nombre de travaux : de la...