Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285201730> ?p ?o ?g. }
- W4285201730 endingPage "53489" @default.
- W4285201730 startingPage "53481" @default.
- W4285201730 abstract "Audio-based automatic speech recognition as a hearing aid is susceptible to background noise and overlapping speeches. Consequently, audio-visual speech recognition has been developed to complement the audio input with additional visual information. However, the huge improvement of neural networks in the visual task has resulted in a robust and reliable lip reading framework that can recognize speech from visual input alone. In this work, we propose a lip reading recognition model to predict daily Mandarin conversation and collect a new Daily Mandarin Conversation Lip Reading (DMCLR) dataset, consisting of 1,000 videos from 100 daily conversations spoken by ten speakers. Our model consists of a spatiotemporal convolution layer, a SE-ResNet-18 network, and a back-end module consisting of bi-directional gated recurrent unit (Bi-GRU), 1D convolution, and fully-connected layers. This model is able to reach 94.2% of accuracy in the DMCLR dataset. Such performance makes it possible for Mandarin lip reading applications to be practical in real life. Additionally, we are able to achieve 86.6% and 57.2% accuracy on Lip Reading in the Wild (LRW) and LRW-1000 (Mandarin), respectively. The results show that our method achieves state-of-the-art performance on these two challenging datasets." @default.
- W4285201730 created "2022-07-14" @default.
- W4285201730 creator A5015972600 @default.
- W4285201730 creator A5026137532 @default.
- W4285201730 creator A5028639489 @default.
- W4285201730 creator A5068251829 @default.
- W4285201730 date "2022-01-01" @default.
- W4285201730 modified "2023-09-25" @default.
- W4285201730 title "Using Lip Reading Recognition to Predict Daily Mandarin Conversation" @default.
- W4285201730 cites W1503933356 @default.
- W4285201730 cites W1526392145 @default.
- W4285201730 cites W2015143272 @default.
- W4285201730 cites W2038952578 @default.
- W4285201730 cites W2060510034 @default.
- W4285201730 cites W2076462394 @default.
- W4285201730 cites W2104095591 @default.
- W4285201730 cites W2113814270 @default.
- W4285201730 cites W2135823751 @default.
- W4285201730 cites W2136155248 @default.
- W4285201730 cites W2152826865 @default.
- W4285201730 cites W2243738093 @default.
- W4285201730 cites W2267805933 @default.
- W4285201730 cites W2289925289 @default.
- W4285201730 cites W2404704342 @default.
- W4285201730 cites W2752782242 @default.
- W4285201730 cites W2893436174 @default.
- W4285201730 cites W2897492880 @default.
- W4285201730 cites W2912581782 @default.
- W4285201730 cites W2938386503 @default.
- W4285201730 cites W2952746495 @default.
- W4285201730 cites W2990152177 @default.
- W4285201730 cites W2999528291 @default.
- W4285201730 cites W3016011581 @default.
- W4285201730 cites W3016232124 @default.
- W4285201730 cites W3034552680 @default.
- W4285201730 cites W3097966334 @default.
- W4285201730 cites W3116228589 @default.
- W4285201730 cites W3128564814 @default.
- W4285201730 cites W3162293946 @default.
- W4285201730 cites W3198127178 @default.
- W4285201730 cites W3199157824 @default.
- W4285201730 cites W4231987518 @default.
- W4285201730 doi "https://doi.org/10.1109/access.2022.3175867" @default.
- W4285201730 hasPublicationYear "2022" @default.
- W4285201730 type Work @default.
- W4285201730 citedByCount "1" @default.
- W4285201730 countsByYear W42852017302023 @default.
- W4285201730 crossrefType "journal-article" @default.
- W4285201730 hasAuthorship W4285201730A5015972600 @default.
- W4285201730 hasAuthorship W4285201730A5026137532 @default.
- W4285201730 hasAuthorship W4285201730A5028639489 @default.
- W4285201730 hasAuthorship W4285201730A5068251829 @default.
- W4285201730 hasBestOaLocation W42852017301 @default.
- W4285201730 hasConcept C115961682 @default.
- W4285201730 hasConcept C138885662 @default.
- W4285201730 hasConcept C138954614 @default.
- W4285201730 hasConcept C153180895 @default.
- W4285201730 hasConcept C154945302 @default.
- W4285201730 hasConcept C15744967 @default.
- W4285201730 hasConcept C162324750 @default.
- W4285201730 hasConcept C17744445 @default.
- W4285201730 hasConcept C187736073 @default.
- W4285201730 hasConcept C199539241 @default.
- W4285201730 hasConcept C204321447 @default.
- W4285201730 hasConcept C23224414 @default.
- W4285201730 hasConcept C2777200299 @default.
- W4285201730 hasConcept C2780451532 @default.
- W4285201730 hasConcept C28490314 @default.
- W4285201730 hasConcept C41008148 @default.
- W4285201730 hasConcept C41895202 @default.
- W4285201730 hasConcept C45347329 @default.
- W4285201730 hasConcept C46312422 @default.
- W4285201730 hasConcept C50644808 @default.
- W4285201730 hasConcept C554936623 @default.
- W4285201730 hasConcept C99498987 @default.
- W4285201730 hasConceptScore W4285201730C115961682 @default.
- W4285201730 hasConceptScore W4285201730C138885662 @default.
- W4285201730 hasConceptScore W4285201730C138954614 @default.
- W4285201730 hasConceptScore W4285201730C153180895 @default.
- W4285201730 hasConceptScore W4285201730C154945302 @default.
- W4285201730 hasConceptScore W4285201730C15744967 @default.
- W4285201730 hasConceptScore W4285201730C162324750 @default.
- W4285201730 hasConceptScore W4285201730C17744445 @default.
- W4285201730 hasConceptScore W4285201730C187736073 @default.
- W4285201730 hasConceptScore W4285201730C199539241 @default.
- W4285201730 hasConceptScore W4285201730C204321447 @default.
- W4285201730 hasConceptScore W4285201730C23224414 @default.
- W4285201730 hasConceptScore W4285201730C2777200299 @default.
- W4285201730 hasConceptScore W4285201730C2780451532 @default.
- W4285201730 hasConceptScore W4285201730C28490314 @default.
- W4285201730 hasConceptScore W4285201730C41008148 @default.
- W4285201730 hasConceptScore W4285201730C41895202 @default.
- W4285201730 hasConceptScore W4285201730C45347329 @default.
- W4285201730 hasConceptScore W4285201730C46312422 @default.
- W4285201730 hasConceptScore W4285201730C50644808 @default.
- W4285201730 hasConceptScore W4285201730C554936623 @default.
- W4285201730 hasConceptScore W4285201730C99498987 @default.
- W4285201730 hasFunder F4320322795 @default.