Objectives The challenge REPERE is part of the objectives of the Content and Interaction Program of the Agence Nationale de la Recherche (ANR), in partnership with the Direction Générale de l’Armement (DGA). REPERE aims at evaluate people recognition within television programs. An evaluation is organized annually with a test in January. Three consortia are funded for a period of 36
Overview Rhythm is an important information for understanding audio data. On both music and speech analysis, the rhythm can help to describe and segment different kind of phenomena. The approach we present led us to propose two new representation : The Rhythm Spectrum and the Tempogram Our approach is not based on any musicological or speech knowledge and aims at finding periodicity in changes
Context Work on features extraction and segmentation carried out in our team provides sets of low-level features or segments that can be considered as basic events. Their exploitation and their combination carried out in different ways can lead to the detection of new features or events of a higher or more semantic level. In previous work, we have studied temporal relationships between basic audio
Groupement d’Intérêt Scientifique PAROLOTHEQUE Par analogie aux tumorothèques, une Parolothèque est une banque d’échantillons de parole enregistrés, obtenus à partir de bilans de trouble de la parole ou du langage ou à partir d’entretiens ou d’interviews de personnes concernées par les pathologies tumorales. Bien que les enregistrements seront orientés par la recherche initiale (ou originelle), une automatisation de la transcription
Project Observatoire des Systèmes d’Indexation et de Recherche d’Information Multimédia (OSIRIM) OSIRIM is a federative project headed by the SAMOVA and SIG research groups, mainly supported by the French government, the Region Midi-Pyrénées and the National Center for the Scientific Research (CNRS). The goal is to propose a homogeneous framework for research works on indexing and information retrieval of multimedia contents. This mainly
Articulatory inversion from audio-visual speech for augmented speech presentation Collaborations CNRS DR6 (Project Coordinator) Grenoble INP CNRS DR1 People involved in SAMOVA team Régine André-Obrecht (scientific coordinator) Jérôme Farinas Support ANR DEFIS program (2008 call) : ANR-08-EMER-001 Scheduled Start time : 1st january 2009 End time : 1st september 2012
Indexing Multimedia Data from Wearable Sensors for diagnostics and treatment of Dementia Problem description With the ageing of population, dementia cases consequently increases in Europe. An early diagnosis prevent insecurity and health worsening in aged people living at home. Generally, dementia is diagnosed in gathering clues of pathological changes in people life. To assess those changes, physicians use neurological examination, neuropsychological
Context This research takes place in a context of audio indexing. After some work on the detection of speech and music, the problem of the position of singing appears. Actually, it is music produced by human voice. In our Speech/Music system, it is classifed mainly in the music category, but it was sometimes taken for speech. The purpose of this work
Context Most of the existing video-search engines rely on context and textual metadata such as the title of the video, tags and comments written by users, etc. In other words, no attempt at understanding the actual content of the video is performed. Content-based audiovisual analysis aims at bridging this so-called semantic gap. Overview Our approach to this problem aims at being as generic,
Context Nearly all methods of audio or video segmentation perform with a priori knowledge. These approaches are based on a spatial-temporal modelling of the content and use decision rules. Currently, it is the only way to reach the semantic quality required by search engines. But only recording collections highly structured, such as broadcast videos of news and sports programmes, and