Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897067191> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W2897067191 abstract "Visual information can improve the performance of automatic speech recognition (ASR), especially in the presence of background noise or different speech modes. A key problem is how to fuse the acoustic and visual features leveraging their complementary information and overcoming the alignment differences between modalities. Current audiovisual ASR (AV-ASR) systems rely on linear interpolation or extrapolation as a pre-processing technique to align audio and visual features, assuming that the feature sequences are aligned frame-by-frame. These pre-processing methods oversimplify the phase difference between lip motion and speech, lacking flexibility and impairing the performance of the system. This paper addresses the fusion of audiovisual features with an alignment neural network (AliNN), relying on recurrent neural network (RNN) with attention model. The proposed front-end model can automatically learn the alignment from the data. The resulting aligned features are concatenated and fed to conventional back-end ASR systems. The proposed front-end system is evaluated with matched and mismatch channel conditions, under clean and noisy recordings. The results show that our proposed approach can relatively outperform the baseline by 24.9% with Gaussian mixture model with hidden Markov model (GMM-HMM) back-end and 2.4% with deep neural network with hidden Markov model (DNN-HMM) back-end." @default.
- W2897067191 created "2018-10-26" @default.
- W2897067191 creator A5008085254 @default.
- W2897067191 creator A5040793194 @default.
- W2897067191 date "2018-07-01" @default.
- W2897067191 modified "2023-10-16" @default.
- W2897067191 title "Aligning Audiovisual Features for Audiovisual Speech Recognition" @default.
- W2897067191 cites W1518556865 @default.
- W2897067191 cites W2071932093 @default.
- W2897067191 cites W2096391593 @default.
- W2897067191 cites W2116258879 @default.
- W2897067191 cites W2121486117 @default.
- W2897067191 cites W2147768505 @default.
- W2897067191 cites W2155765376 @default.
- W2897067191 cites W2157190406 @default.
- W2897067191 cites W2397098974 @default.
- W2897067191 cites W2406846463 @default.
- W2897067191 cites W2491224255 @default.
- W2897067191 cites W2509591453 @default.
- W2897067191 cites W2604379605 @default.
- W2897067191 cites W2737658251 @default.
- W2897067191 cites W2746799361 @default.
- W2897067191 cites W2749694333 @default.
- W2897067191 cites W2790326622 @default.
- W2897067191 cites W4238319993 @default.
- W2897067191 doi "https://doi.org/10.1109/icme.2018.8486455" @default.
- W2897067191 hasPublicationYear "2018" @default.
- W2897067191 type Work @default.
- W2897067191 sameAs 2897067191 @default.
- W2897067191 citedByCount "22" @default.
- W2897067191 countsByYear W28970671912018 @default.
- W2897067191 countsByYear W28970671912019 @default.
- W2897067191 countsByYear W28970671912020 @default.
- W2897067191 countsByYear W28970671912021 @default.
- W2897067191 countsByYear W28970671912022 @default.
- W2897067191 countsByYear W28970671912023 @default.
- W2897067191 crossrefType "proceedings-article" @default.
- W2897067191 hasAuthorship W2897067191A5008085254 @default.
- W2897067191 hasAuthorship W2897067191A5040793194 @default.
- W2897067191 hasConcept C111919701 @default.
- W2897067191 hasConcept C119599485 @default.
- W2897067191 hasConcept C126042441 @default.
- W2897067191 hasConcept C127413603 @default.
- W2897067191 hasConcept C138885662 @default.
- W2897067191 hasConcept C141353440 @default.
- W2897067191 hasConcept C153180895 @default.
- W2897067191 hasConcept C154945302 @default.
- W2897067191 hasConcept C23224414 @default.
- W2897067191 hasConcept C2776401178 @default.
- W2897067191 hasConcept C28490314 @default.
- W2897067191 hasConcept C36464697 @default.
- W2897067191 hasConcept C41008148 @default.
- W2897067191 hasConcept C41895202 @default.
- W2897067191 hasConcept C50644808 @default.
- W2897067191 hasConcept C53016008 @default.
- W2897067191 hasConcept C61328038 @default.
- W2897067191 hasConcept C76155785 @default.
- W2897067191 hasConceptScore W2897067191C111919701 @default.
- W2897067191 hasConceptScore W2897067191C119599485 @default.
- W2897067191 hasConceptScore W2897067191C126042441 @default.
- W2897067191 hasConceptScore W2897067191C127413603 @default.
- W2897067191 hasConceptScore W2897067191C138885662 @default.
- W2897067191 hasConceptScore W2897067191C141353440 @default.
- W2897067191 hasConceptScore W2897067191C153180895 @default.
- W2897067191 hasConceptScore W2897067191C154945302 @default.
- W2897067191 hasConceptScore W2897067191C23224414 @default.
- W2897067191 hasConceptScore W2897067191C2776401178 @default.
- W2897067191 hasConceptScore W2897067191C28490314 @default.
- W2897067191 hasConceptScore W2897067191C36464697 @default.
- W2897067191 hasConceptScore W2897067191C41008148 @default.
- W2897067191 hasConceptScore W2897067191C41895202 @default.
- W2897067191 hasConceptScore W2897067191C50644808 @default.
- W2897067191 hasConceptScore W2897067191C53016008 @default.
- W2897067191 hasConceptScore W2897067191C61328038 @default.
- W2897067191 hasConceptScore W2897067191C76155785 @default.
- W2897067191 hasLocation W28970671911 @default.
- W2897067191 hasOpenAccess W2897067191 @default.
- W2897067191 hasPrimaryLocation W28970671911 @default.
- W2897067191 hasRelatedWork W1542012215 @default.
- W2897067191 hasRelatedWork W2008638795 @default.
- W2897067191 hasRelatedWork W2150288981 @default.
- W2897067191 hasRelatedWork W2161510337 @default.
- W2897067191 hasRelatedWork W2382607599 @default.
- W2897067191 hasRelatedWork W2539985974 @default.
- W2897067191 hasRelatedWork W2546942002 @default.
- W2897067191 hasRelatedWork W2897067191 @default.
- W2897067191 hasRelatedWork W2908073754 @default.
- W2897067191 hasRelatedWork W2970216048 @default.
- W2897067191 isParatext "false" @default.
- W2897067191 isRetracted "false" @default.
- W2897067191 magId "2897067191" @default.
- W2897067191 workType "article" @default.