Speaker diarization de fichiers vidéos hétérogènes issus du web


In the last ten years, Internet changed significantly. The main change is certainly the content of the Internet, in its quantity, its variety or the media used to show it. Regarding multimedia, the most impressive evolution is the continuous growing success of the video sharing websites. But, with this success come the difficulties to efficiently search, index and access relevant information about these documents. Speaker diarization is an important task in the overall information retrieval process. This paper describes the audio/video database, especially built for the speaker diarization task, based on different video genres. Through some preliminary experiments, it highlights the difficulties encountered in this context, mainly linked to the database heterogeneity. MOTS-CLÉS : segmentation-regroupement en locuteurs, vidéos hétérogènes, erreur de diarization

DOI: 10.24348/coria.2011.379

Extracted Key Phrases

Cite this paper

@inproceedings{Clment2011SpeakerDD, title={Speaker diarization de fichiers vid{\'e}os h{\'e}t{\'e}rog{\`e}nes issus du web}, author={Pierre Cl{\'e}ment}, booktitle={CORIA}, year={2011} }