https:\/\/meet.google.com\/egw-ienb-bch<\/a><\/p>\n\n\n\nMembres du jury :<\/strong><\/p>\n\n\n\nM. Bj\u00f6rn SCHULLER<\/strong>, Professeur, Imperial College London, ROYAUME-UNI – Rapporteur
M. Sylvain MEIGNIER<\/strong>, Professeur des universit\u00e9s, Laboratoire d\u2019Informatique de l\u2019Universit\u00e9 du Mans, FRANCE – Rapporteur
Mme Dijana PETROVSKA<\/strong>, Ma\u00eetre de conf\u00e9rences, Telecom SudParis, FRANCE – Examinateur
M. J\u00e9r\u00f4me BOUDY<\/strong>, Professeur, Telecom SudParis, FRANCE – Directeur de these
M. G\u00e9rard CHOLLET<\/strong>, Directeur de recherche \u00e9m\u00e9rite, Telecom SudParis, FRANCE – CoDirecteur de these
M. Herv\u00e9 BREDIN<\/strong>, Charg\u00e9 de recherche, IRIT, FRANCE – Examinateur<\/p>\n\n\n\n\u00ab Diarisation multimodale : vers des mod\u00e8les robustes et justes en contexte r\u00e9el \u00bb<\/h2>\n\n\n\npr\u00e9sent\u00e9 par Monsieur Yannis TEVISSEN<\/h2>\n\n\n\n
R\u00e9sum\u00e9 :<\/strong><\/p>\n\n\n\nLa diarisation du locuteur, c\u2019est \u00e0 dire la tache de d\u00e9terminer automatiquement \u00ab qui parle, quand ? \u00bb dans un enregistrement audio ou vid\u00e9o, est un des piliers des syst\u00e8mes modernes d\u2019analyse des conversations. A la t\u00e9l\u00e9vision, les contenus diffus\u00e9s sont divers et couvrent \u00e0 peu pr\u00e8s tous les types de conversations, de la discussion calme entre deux personnes aux d\u00e9bats passionn\u00e9s en passant par les interviews en terrain de guerre. L\u2019archivage et l\u2019indexation de ces contenus, r\u00e9alis\u00e9s par la soci\u00e9t\u00e9 Newsbridge, requi\u00e8rent des m\u00e9thodes de traitement robustes et justes pour les analyser. Dans ce travail, nous pr\u00e9sentons deux nouvelles m\u00e9thodes permettant d\u2019am\u00e9liorer la robustesse des syst\u00e8mes via des approches de fusion. La premi\u00e8re se concentre sur la d\u00e9tection d\u2019activit\u00e9 vocale, pr\u00e9traitement n\u00e9cessaire \u00e0 tout syst\u00e8me de diarisation. La seconde est une approche multimodale qui tire notamment parti des derni\u00e8res avanc\u00e9es en traitement du langage naturel. Nous voyons \u00e9galement que les r\u00e9centes avanc\u00e9es des syst\u00e8mes de diarisation rendent l\u2019utilisation de la diarisation du locuteur r\u00e9aliste y compris dans des secteurs critiques tels que l\u2019analyse de larges archives audiovisuelles ou le maintien \u00e0 domicile de personnes \u00e2g\u00e9es. Enfin ce travail pr\u00e9sente une nouvelle m\u00e9thode d\u2019\u00e9valuation de la justesse algorithmique de la diarisation du locuteur en vue de rendre son utilisation plus responsable.<\/p>\n\n\n\n
Abstract :<\/strong><\/p>\n\n\n\nSpeaker diarization, or the task of automatically determining \u00ab\u00a0who spoke, when?\u00a0\u00bb in an audio or video recording, is one of the pillars of modern conversation analysis systems. On television, the content broadcasted are very diverse and covers about every type of conversation, from calm discussions between two people to impassioned debates and wartime interviews. The archiving and indexing of this content, carried out by the Newsbridge company, requires robust and fair processing methods to analyze them. In this work, we present two new methods for improving systems robustness via fusion approaches. The first method focuses on voice activity detection, a necessary pre-processing step for every diarization system. The second is a multimodal approach that takes advantage of the latest advances in natural language processing. We also show that recent advances in diarization systems make the use of speaker diarization realistic, even in critical sectors such as the analysis of large audiovisual archives or the home care of the elderly. Finally, this work shows a new method for evaluating the algorithmic fairness of speaker diarization, with the objective to make its use more responsible.<\/p>\n","protected":false},"excerpt":{"rendered":"
L’Ecole doctorale : Ecole Doctorale de l’Institut Polytechnique de Paris et le Laboratoire de recherche SAMOVAR – Services r\u00e9partis, Architectures, Mod\u00e9lisation, Validation, Administration des R\u00e9seaux pr\u00e9sentent l\u2019AVIS DE SOUTENANCE de Monsieur Yannis TEVISSEN Autoris\u00e9 \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en : […]<\/p>\n","protected":false},"author":4,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ocean_post_layout":"","ocean_both_sidebars_style":"","ocean_both_sidebars_content_width":0,"ocean_both_sidebars_sidebars_width":0,"ocean_sidebar":"0","ocean_second_sidebar":"0","ocean_disable_margins":"enable","ocean_add_body_class":"","ocean_shortcode_before_top_bar":"","ocean_shortcode_after_top_bar":"","ocean_shortcode_before_header":"","ocean_shortcode_after_header":"","ocean_has_shortcode":"","ocean_shortcode_after_title":"","ocean_shortcode_before_footer_widgets":"","ocean_shortcode_after_footer_widgets":"","ocean_shortcode_before_footer_bottom":"","ocean_shortcode_after_footer_bottom":"","ocean_display_top_bar":"default","ocean_display_header":"default","ocean_header_style":"","ocean_center_header_left_menu":"0","ocean_custom_header_template":"0","ocean_custom_logo":0,"ocean_custom_retina_logo":0,"ocean_custom_logo_max_width":0,"ocean_custom_logo_tablet_max_width":0,"ocean_custom_logo_mobile_max_width":0,"ocean_custom_logo_max_height":0,"ocean_custom_logo_tablet_max_height":0,"ocean_custom_logo_mobile_max_height":0,"ocean_header_custom_menu":"0","ocean_menu_typo_font_family":"0","ocean_menu_typo_font_subset":"","ocean_menu_typo_font_size":0,"ocean_menu_typo_font_size_tablet":0,"ocean_menu_typo_font_size_mobile":0,"ocean_menu_typo_font_size_unit":"px","ocean_menu_typo_font_weight":"","ocean_menu_typo_font_weight_tablet":"","ocean_menu_typo_font_weight_mobile":"","ocean_menu_typo_transform":"","ocean_menu_typo_transform_tablet":"","ocean_menu_typo_transform_mobile":"","ocean_menu_typo_line_height":0,"ocean_menu_typo_line_height_tablet":0,"ocean_menu_typo_line_height_mobile":0,"ocean_menu_typo_line_height_unit":"","ocean_menu_typo_spacing":0,"ocean_menu_typo_spacing_tablet":0,"ocean_menu_typo_spacing_mobile":0,"ocean_menu_typo_spacing_unit":"","ocean_menu_link_color":"","ocean_menu_link_color_hover":"","ocean_menu_link_color_active":"","ocean_menu_link_background":"","ocean_menu_link_hover_background":"","ocean_menu_link_active_background":"","ocean_menu_social_links_bg":"","ocean_menu_social_hover_links_bg":"","ocean_menu_social_links_color":"","ocean_menu_social_hover_links_color":"","ocean_disable_title":"default","ocean_disable_heading":"default","ocean_post_title":"","ocean_post_subheading":"","ocean_post_title_style":"","ocean_post_title_background_color":"","ocean_post_title_background":0,"ocean_post_title_bg_image_position":"","ocean_post_title_bg_image_attachment":"","ocean_post_title_bg_image_repeat":"","ocean_post_title_bg_image_size":"","ocean_post_title_height":0,"ocean_post_title_bg_overlay":0.5,"ocean_post_title_bg_overlay_color":"","ocean_disable_breadcrumbs":"default","ocean_breadcrumbs_color":"","ocean_breadcrumbs_separator_color":"","ocean_breadcrumbs_links_color":"","ocean_breadcrumbs_links_hover_color":"","ocean_display_footer_widgets":"default","ocean_display_footer_bottom":"default","ocean_custom_footer_template":"0","ocean_post_oembed":"","ocean_post_self_hosted_media":"","ocean_post_video_embed":"","ocean_link_format":"","ocean_link_format_target":"self","ocean_quote_format":"","ocean_quote_format_link":"post","ocean_gallery_link_images":"off","ocean_gallery_id":[],"footnotes":""},"categories":[286,169],"tags":[],"class_list":["post-6399","post","type-post","status-publish","format-standard","hentry","category-fractualites-ennews-fr","category-seminaires-armedia","entry"],"_links":{"self":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6399","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/comments?post=6399"}],"version-history":[{"count":1,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6399\/revisions"}],"predecessor-version":[{"id":6400,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6399\/revisions\/6400"}],"wp:attachment":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/media?parent=6399"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/categories?post=6399"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/tags?post=6399"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}