{"id":82797,"date":"2000-01-01T00:00:00","date_gmt":"2000-01-01T00:00:00","guid":{"rendered":"https:\/\/www.deberes.net\/tesis\/sin-categoria\/improved-modelling-for-robust-speech-recognition\/"},"modified":"2000-01-01T00:00:00","modified_gmt":"2000-01-01T00:00:00","slug":"improved-modelling-for-robust-speech-recognition","status":"publish","type":"post","link":"https:\/\/www.deberes.net\/tesis\/ciencias-tecnologicas\/improved-modelling-for-robust-speech-recognition\/","title":{"rendered":"Improved modelling for robust speech recognition."},"content":{"rendered":"<h2>Tesis doctoral de <strong> Pau Paches Leal <\/strong><\/h2>\n<p>Una de las l\u00edneas seguidas en esta tesis es intentar conocer mejor nuevas estrategias para mejorar el reconocimiento del habla. En este trabajo, se presenta un nuevo algoritmo (mca) para compensar las inhomogeneidades en el dominio del espectro de modulaci\u00f3n, que tiene cierto sentido perceptual y en el que se pueden representar las variaciones temporales de la se\u00f1al. Mca es un procedimiento de m\u00e1xima verosimilitud para la estimaci\u00f3n autom\u00e1tica de filtros en el espectro de modulaci\u00f3n, de cara a compensar distorsiones en este dominio.  dos bases de datos de prop\u00f3sito general, speechdat espa\u00f1ola y speechdat catalana, se usan en este trabajo. La modelizaci\u00f3n independiente de la tarea, que consiste en entrenar modelos fon\u00e9ticos generales a partir de frases equilibradas fon\u00e9ticamente, es la estrategia usada aqu\u00ed. se lleva a cabo un estudio sobre las unidades pr\u00e1cticas para crear sistemas independientes de la tarea de tama\u00f1o mediano. Unidades m\u00e1s sencillas que hacen suposiciones simplificadoras sobre los efectos del contexto se comparan con los muy conocidos trifonemas. M\u00e9todos de ligadura de estados basados en \u00e1rboles de decisi\u00f3n se usan ampliamente aqu\u00ed para hacer entrenables las unidades dependientes del contexto usadas. Se efect\u00faan dos estudios independientes, uno para un sistema de reconocimiento en castellano y el otro para un sistema de reconocimiento en catal\u00e1n.  un diccionario fon\u00e9tico se necesita para entrenar un sistema de reconocimiento basado en unidades subl\u00e9xicas. La obtenci\u00f3n de un diccionario fon\u00e9tico es muy costosa en tiempo. Un conversor autom\u00e1tico grafema-fonema, segre, para la lengua catalana ha sido desarrollado en el marco de esta tesis y se ha usado para construir sistemas de reconocimiento en catal\u00e1n para la base speechdat. La caracter\u00edstica principal de este transcriptor es que las reglas de conversi\u00f3n no est\u00e1n fijas dentro del c\u00f3digo del programa sino que s<\/p>\n<p>&nbsp;<\/p>\n<h3>Datos acad\u00e9micos de la tesis doctoral \u00ab<strong>Improved modelling for robust speech recognition.<\/strong>\u00ab<\/h3>\n<ul>\n<li><strong>T\u00edtulo de la tesis:<\/strong>\u00a0 Improved modelling for robust speech recognition. <\/li>\n<li><strong>Autor:<\/strong>\u00a0 Pau Paches Leal <\/li>\n<li><strong>Universidad:<\/strong>\u00a0 Polit\u00e9cnica de catalunya<\/li>\n<li><strong>Fecha de lectura de la tesis:<\/strong>\u00a0 01\/01\/2000<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h3>Direcci\u00f3n y tribunal<\/h3>\n<ul>\n<li><strong>Director de la tesis<\/strong>\n<ul>\n<li>Climent Nadeu Camprubi<\/li>\n<\/ul>\n<\/li>\n<li><strong>Tribunal<\/strong>\n<ul>\n<li>Presidente del tribunal: Jos\u00e9 bernardo Mari\u00f1o acebal <\/li>\n<li>joaquim Llisterri boix (vocal)<\/li>\n<li>horacio Rodr\u00edguez hontoria (vocal)<\/li>\n<li>thierry Dutoit (vocal)<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Tesis doctoral de Pau Paches Leal Una de las l\u00edneas seguidas en esta tesis es intentar conocer mejor nuevas estrategias [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""}},"footnotes":""},"categories":[332,15596,27844,2489,2535],"tags":[30527,13269,43676,17031,175841,175842],"class_list":["post-82797","post","type-post","status-publish","format-standard","hentry","category-ciencias-tecnologicas","category-politecnica-de-catalunya","category-reconocimiento-y-sintetizacion-de-habla","category-tecnologia-de-las-telecomunicaciones","category-tecnologia-de-los-ordenadores","tag-climent-nadeu-camprubi","tag-horacio-rodriguez-hontoria","tag-joaquim-llisterri-boix","tag-jose-bernardo-marino-acebal","tag-pau-paches-leal","tag-thierry-dutoit"],"_links":{"self":[{"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/posts\/82797","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/comments?post=82797"}],"version-history":[{"count":0,"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/posts\/82797\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/media?parent=82797"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/categories?post=82797"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.deberes.net\/tesis\/wp-json\/wp\/v2\/tags?post=82797"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}