Matches in SemOpenAlex for { <https://semopenalex.org/work/W3020804303> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3020804303 endingPage "285" @default.
- W3020804303 startingPage "273" @default.
- W3020804303 abstract "Abstract Noisy situations cause huge problems for the hearing-impaired, as hearing aids often make speech more audible but do not always restore intelligibility. In noisy settings, humans routinely exploit the audio-visual (AV) nature of speech to selectively suppress background noise and focus on the target speaker. In this paper, we present a novel language-, noise- and speaker-independent AV deep neural network (DNN) architecture, termed CochleaNet, for causal or real-time speech enhancement (SE). The model jointly exploits noisy acoustic cues and noise robust visual cues to focus on the desired speaker and improve speech intelligibility. The proposed SE framework is evaluated using a first of its kind AV binaural speech corpus, ASPIRE, recorded in real noisy environments, including cafeteria and restaurant settings. We demonstrate superior performance of our approach in terms of both objective measures and subjective listening tests, over state-of-the-art SE approaches, including recent DNN based SE models. In addition, our work challenges a popular belief that scarcity of a multi-lingual, large vocabulary AV corpus and a wide variety of noises is a major bottleneck to build robust language, speaker and noise-independent SE systems. We show that a model trained on a synthetic mixture of the benchmark GRID corpus (with 33 speakers and a small English vocabulary) and CHiME 3 noises (comprising bus, pedestrian, cafeteria, and street noises) can generalise well, not only on large vocabulary corpora with a wide variety of speakers and noises, but also on completely unrelated languages such as Mandarin." @default.
- W3020804303 created "2020-05-01" @default.
- W3020804303 creator A5058869522 @default.
- W3020804303 creator A5062211930 @default.
- W3020804303 creator A5068981769 @default.
- W3020804303 creator A5082626629 @default.
- W3020804303 date "2020-11-01" @default.
- W3020804303 modified "2023-09-29" @default.
- W3020804303 title "CochleaNet: A robust language-independent audio-visual model for real-time speech enhancement" @default.
- W3020804303 cites W1974387177 @default.
- W3020804303 cites W2010929291 @default.
- W3020804303 cites W2015143272 @default.
- W3020804303 cites W2015394094 @default.
- W3020804303 cites W2021279213 @default.
- W3020804303 cites W2027701650 @default.
- W3020804303 cites W2029199293 @default.
- W3020804303 cites W2055516313 @default.
- W3020804303 cites W2064675550 @default.
- W3020804303 cites W2069681747 @default.
- W3020804303 cites W2081144555 @default.
- W3020804303 cites W2120847449 @default.
- W3020804303 cites W2121973264 @default.
- W3020804303 cites W2141998673 @default.
- W3020804303 cites W2144404214 @default.
- W3020804303 cites W2740594650 @default.
- W3020804303 cites W2788241093 @default.
- W3020804303 cites W2886232760 @default.
- W3020804303 cites W2888868298 @default.
- W3020804303 cites W2912227124 @default.
- W3020804303 cites W2921691172 @default.
- W3020804303 cites W2946704636 @default.
- W3020804303 cites W2952218014 @default.
- W3020804303 cites W2952979574 @default.
- W3020804303 cites W2962866211 @default.
- W3020804303 cites W2963341071 @default.
- W3020804303 cites W2971630540 @default.
- W3020804303 cites W2972177675 @default.
- W3020804303 cites W4233392025 @default.
- W3020804303 cites W4289665794 @default.
- W3020804303 doi "https://doi.org/10.1016/j.inffus.2020.04.001" @default.
- W3020804303 hasPublicationYear "2020" @default.
- W3020804303 type Work @default.
- W3020804303 sameAs 3020804303 @default.
- W3020804303 citedByCount "51" @default.
- W3020804303 countsByYear W30208043032019 @default.
- W3020804303 countsByYear W30208043032020 @default.
- W3020804303 countsByYear W30208043032021 @default.
- W3020804303 countsByYear W30208043032022 @default.
- W3020804303 countsByYear W30208043032023 @default.
- W3020804303 crossrefType "journal-article" @default.
- W3020804303 hasAuthorship W3020804303A5058869522 @default.
- W3020804303 hasAuthorship W3020804303A5062211930 @default.
- W3020804303 hasAuthorship W3020804303A5068981769 @default.
- W3020804303 hasAuthorship W3020804303A5082626629 @default.
- W3020804303 hasBestOaLocation W30208043032 @default.
- W3020804303 hasConcept C154945302 @default.
- W3020804303 hasConcept C163294075 @default.
- W3020804303 hasConcept C2776182073 @default.
- W3020804303 hasConcept C28490314 @default.
- W3020804303 hasConcept C3017588708 @default.
- W3020804303 hasConcept C41008148 @default.
- W3020804303 hasConcept C49774154 @default.
- W3020804303 hasConceptScore W3020804303C154945302 @default.
- W3020804303 hasConceptScore W3020804303C163294075 @default.
- W3020804303 hasConceptScore W3020804303C2776182073 @default.
- W3020804303 hasConceptScore W3020804303C28490314 @default.
- W3020804303 hasConceptScore W3020804303C3017588708 @default.
- W3020804303 hasConceptScore W3020804303C41008148 @default.
- W3020804303 hasConceptScore W3020804303C49774154 @default.
- W3020804303 hasFunder F4320334627 @default.
- W3020804303 hasLocation W30208043031 @default.
- W3020804303 hasLocation W30208043032 @default.
- W3020804303 hasLocation W30208043033 @default.
- W3020804303 hasOpenAccess W3020804303 @default.
- W3020804303 hasPrimaryLocation W30208043031 @default.
- W3020804303 hasRelatedWork W2577762507 @default.
- W3020804303 hasRelatedWork W2794908863 @default.
- W3020804303 hasRelatedWork W3004146833 @default.
- W3020804303 hasRelatedWork W3015612952 @default.
- W3020804303 hasRelatedWork W3081997829 @default.
- W3020804303 hasRelatedWork W3096028031 @default.
- W3020804303 hasRelatedWork W3146437239 @default.
- W3020804303 hasRelatedWork W4298300178 @default.
- W3020804303 hasRelatedWork W4300529166 @default.
- W3020804303 hasRelatedWork W4304891817 @default.
- W3020804303 hasVolume "63" @default.
- W3020804303 isParatext "false" @default.
- W3020804303 isRetracted "false" @default.
- W3020804303 magId "3020804303" @default.
- W3020804303 workType "article" @default.