{"id":1696,"date":"2022-09-04T03:44:31","date_gmt":"2022-09-04T01:44:31","guid":{"rendered":"https:\/\/www-staging.systransoft.com\/papers\/incidunt-quia-magni-ab\/"},"modified":"2023-11-16T18:22:18","modified_gmt":"2023-11-16T17:22:18","slug":"incidunt-quia-magni-ab","status":"publish","type":"paper","link":"https:\/\/www-staging.systransoft.com\/es\/recursos\/papers-and-publications\/incidunt-quia-magni-ab\/","title":{"rendered":"Traducci\u00f3n robusta de transcripciones de habla en vivo en franc\u00e9s"},"content":{"rendered":"Despite a narrowed performance gap with direct approaches, cascade solutions, involving automatic speech recognition (ASR) and machine translation (MT) are still largely employed in speech translation (ST). Direct approaches employing a single model to translate the input speech signal suffer from the critical bottleneck of data scarcity. In addition, multiple industry applications display speech transcripts alongside translations, making cascade approaches more realistic and practical. In the context of cascaded simultaneous ST, we propose several solutions to adapt a neural MT network to take as input the transcripts output by an ASR system. Adaptation is achieved by enriching speech transcripts and MT data sets so that they more closely resemble each other, thereby improving the system robustness to error propagation and enhancing result legibility for humans. We address aspects such as sentence boundaries, capitalisation, punctuation, hesitations, repetitions, homophones, etc. while taking into account the low latency requirement of simultaneous ST systems.","protected":false},"excerpt":{"rendered":"<p>A pesar de la reducci\u00f3n de la brecha de rendimiento con los enfoques directos, las soluciones en cascada, que implican el reconocimiento autom\u00e1tico de voz (ASR) y la traducci\u00f3n autom\u00e1tica (MT) todav\u00eda se emplean en gran medida en la traducci\u00f3n de voz (ST). Los enfoques directos que emplean un \u00fanico modelo para traducir la se\u00f1al de entrada de voz sufren del cuello de botella cr\u00edtico de la escasez de datos. Adem\u00e1s, m\u00faltiples aplicaciones de la industria muestran transcripciones de voz ... <a href=\"https:\/\/www-staging.systransoft.com\/es\/recursos\/papers-and-publications\/incidunt-quia-magni-ab\/\">Continuaci\u00f3n<\/a><\/p>","protected":false},"featured_media":0,"template":"","class_list":["post-1696","paper","type-paper","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/www-staging.systransoft.com\/es\/wp-json\/wp\/v2\/paper\/1696","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www-staging.systransoft.com\/es\/wp-json\/wp\/v2\/paper"}],"about":[{"href":"https:\/\/www-staging.systransoft.com\/es\/wp-json\/wp\/v2\/types\/paper"}],"wp:attachment":[{"href":"https:\/\/www-staging.systransoft.com\/es\/wp-json\/wp\/v2\/media?parent=1696"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}