{"id":320,"date":"2017-10-26T15:04:30","date_gmt":"2017-10-26T13:04:30","guid":{"rendered":"http:\/\/members.loria.fr\/LPerotin\/?page_id=320"},"modified":"2019-04-09T11:24:38","modified_gmt":"2019-04-09T09:24:38","slug":"icassp2018","status":"publish","type":"page","link":"https:\/\/members.loria.fr\/LPerotin\/demos\/icassp2018\/","title":{"rendered":"Speech separation"},"content":{"rendered":"<h1>Situation with a single speaker and diffuse crowd noise<\/h1>\n<h3>Initial clean speech:<\/h3>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-1\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_dry.wav?_=1\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_dry.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_dry.wav<\/a><\/audio>\n<\/div>\n<h3>Mixture with reverberation and diffuse noise at 0 dB SNR:<\/h3>\n<ul>\n<li>3D spatial scene made with binaural rendering (listen with headphones):<\/li>\n<\/ul>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-2\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_xHoa.wav?_=2\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_xHoa.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_xHoa.wav<\/a><\/audio>\n<\/div>\n<ul>\n<li>Mono signal:<\/li>\n<\/ul>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-3\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_x.wav?_=3\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_x.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_x.wav<\/a><\/audio>\n<\/div>\n<h3>Speech enhanced with a simple directional filter (delay-and-sum-type beamformer):<\/h3>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-4\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_ds.wav?_=4\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_ds.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_ds.wav<\/a><\/audio>\n<\/div>\n<h3>Speech enhanced with our LSTM-based system:<\/h3>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-5\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_omniBeamDiff_guess.wav?_=5\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_omniBeamDiff_guess.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_noise_omniBeamDiff_guess.wav<\/a><\/audio>\n<\/div>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n<h1>Situation with two speakers (25\u00b0apart) and diffuse crowd noise<\/h1>\n<h3>Initial clean speech:<\/h3>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-6\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_dry.wav?_=6\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_dry.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_dry.wav<\/a><\/audio>\n<\/div>\n<h3>Mixture with reverberation, a competing speaker at 0 dB SIR and diffuse noise at 20 dB SNR:<\/h3>\n<ul>\n<li>3D spatial scene made with binaural rendering (listen with headphones):<\/li>\n<\/ul>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-7\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_xHoa.wav?_=7\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_xHoa.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_xHoa.wav<\/a><\/audio>\n<\/div>\n<ul>\n<li>Mono signal:<\/li>\n<\/ul>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-8\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_x.wav?_=8\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_x.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_x.wav<\/a><\/audio>\n<\/div>\n<h3>Speech enhanced with a simple directional filter (delay-and-sum-type beamformer):<\/h3>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-9\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_ds.wav?_=9\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_ds.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_ds.wav<\/a><\/audio>\n<\/div>\n<h3>Speech enhanced with our LSTM-based system:<\/h3>\n<div style=\"width: 50%\">\n<audio class=\"wp-audio-shortcode\" id=\"audio-320-10\" preload=\"none\" style=\"width: 100%;\" controls=\"controls\"><source type=\"audio\/wav\" src=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_omniBeamPonct_guess.wav?_=10\" \/><a href=\"\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_omniBeamPonct_guess.wav\">\/LPerotin\/files\/audio\/Ester_AR_00_00_speech_dAngle025_omniBeamPonct_guess.wav<\/a><\/audio>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Situation with a single speaker and diffuse crowd noise<br \/>\nInitial clean speech:<\/p>\n<p>Mixture with reverberation and diffuse noise at 0 dB SNR:<\/p>\n<ul>\n<li>3D spatial scene made with binaural rendering (listen with headphones):<\/li>\n<\/ul>\n<ul>\n<li>Mono signal:<\/li>\n<\/ul>\n<p>Speech enhanced with a simple directional filter (delay-and-sum-type beamformer):<\/p>\n<p>Speech enhanced with our LSTM-based system:<\/p>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n<p>Situation with two speakers (25\u00b0apart) and diffuse crowd noise<br \/>\nInitial clean speech:<\/p>\n<p>Mixture with reverberation, a competing speaker at 0 dB SIR and diffuse noise at 20 dB SNR:<\/p>\n<ul>\n<li>3D spatial scene made with binaural rendering (listen with headphones):<\/li>\n<\/ul>\n<ul>\n<li>Mono signal:<\/li>\n<\/ul>\n<p>Speech enhanced with a simple directional filter (delay-and-sum-type beamformer):<\/p>\n<p>Speech enhanced with our LSTM-based system: <\/p>\n","protected":false},"author":157,"featured_media":0,"parent":343,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-320","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/pages\/320","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/users\/157"}],"replies":[{"embeddable":true,"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/comments?post=320"}],"version-history":[{"count":21,"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/pages\/320\/revisions"}],"predecessor-version":[{"id":394,"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/pages\/320\/revisions\/394"}],"up":[{"embeddable":true,"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/pages\/343"}],"wp:attachment":[{"href":"https:\/\/members.loria.fr\/LPerotin\/wp-json\/wp\/v2\/media?parent=320"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}