FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3024659761> ?p ?o ?g. }

• W3024659761 abstract "Automatic speech recognition (ASR) of overlapped speech remains a highly challenging task to date. To this end, multi-channel microphone array data are widely used in state-of-the-art ASR systems. Motivated by the invariance of visual modality to acoustic signal corruption, this paper presents an audio-visual multi-channel overlapped speech recognition system featuring tightly integrated separation front-end and recognition back-end. A series of audio-visual multi-channel speech separation front-end components based on textit{TF masking}, textit{filter&sum} and textit{mask-based MVDR} beamforming approaches were developed. To reduce the error cost mismatch between the separation and recognition components, they were jointly fine-tuned using the connectionist temporal classification (CTC) loss function, or a multi-task criterion interpolation with scale-invariant signal to noise ratio (Si-SNR) error cost. Experiments suggest that the proposed multi-channel AVSR system outperforms the baseline audio-only ASR system by up to 6.81% (26.83% relative) and 22.22% (56.87% relative) absolute word error rate (WER) reduction on overlapped speech constructed using either simulation or replaying of the lipreading sentence 2 (LRS2) dataset respectively." @default.
• W3024659761 created "2020-05-21" @default.
• W3024659761 creator A5004643540 @default.
• W3024659761 creator A5019458385 @default.
• W3024659761 creator A5034476404 @default.
• W3024659761 creator A5035661649 @default.
• W3024659761 creator A5037109470 @default.
• W3024659761 creator A5038895203 @default.
• W3024659761 creator A5056567731 @default.
• W3024659761 creator A5057987365 @default.
• W3024659761 creator A5072495935 @default.
• W3024659761 creator A5075183307 @default.
• W3024659761 date "2020-05-18" @default.
• W3024659761 modified "2023-10-14" @default.
• W3024659761 title "Audio-visual Multi-channel Recognition of Overlapped Speech" @default.
• W3024659761 cites W2042860487 @default.
• W3024659761 cites W2060108923 @default.
• W3024659761 cites W2102764819 @default.
• W3024659761 cites W2148613904 @default.
• W3024659761 cites W2158143227 @default.
• W3024659761 cites W2288645994 @default.
• W3024659761 cites W2291877678 @default.
• W3024659761 cites W2398042854 @default.
• W3024659761 cites W2398972335 @default.
• W3024659761 cites W2508092598 @default.
• W3024659761 cites W2517616541 @default.
• W3024659761 cites W2589857635 @default.
• W3024659761 cites W2640112133 @default.
• W3024659761 cites W2803322398 @default.
• W3024659761 cites W2889503488 @default.
• W3024659761 cites W2890244912 @default.
• W3024659761 cites W2890952074 @default.
• W3024659761 cites W2909607850 @default.
• W3024659761 cites W2939360348 @default.
• W3024659761 cites W2939497224 @default.
• W3024659761 cites W2952218014 @default.
• W3024659761 cites W2963290645 @default.
• W3024659761 cites W2964171275 @default.
• W3024659761 cites W2964283370 @default.
• W3024659761 cites W2972413216 @default.
• W3024659761 cites W2972513594 @default.
• W3024659761 cites W2972693890 @default.
• W3024659761 cites W2973071728 @default.
• W3024659761 cites W2973231102 @default.
• W3024659761 cites W2987346570 @default.
• W3024659761 cites W2995166068 @default.
• W3024659761 cites W3004309045 @default.
• W3024659761 cites W3008400075 @default.
• W3024659761 cites W3011424113 @default.
• W3024659761 cites W3015372568 @default.
• W3024659761 cites W3015383493 @default.
• W3024659761 cites W3015834770 @default.
• W3024659761 cites W3097173744 @default.
• W3024659761 cites W3097909406 @default.
• W3024659761 doi "https://doi.org/10.48550/arxiv.2005.08571" @default.
• W3024659761 hasPublicationYear "2020" @default.
• W3024659761 type Work @default.
• W3024659761 sameAs 3024659761 @default.
• W3024659761 citedByCount "1" @default.
• W3024659761 countsByYear W30246597612020 @default.
• W3024659761 crossrefType "posted-content" @default.
• W3024659761 hasAuthorship W3024659761A5004643540 @default.
• W3024659761 hasAuthorship W3024659761A5019458385 @default.
• W3024659761 hasAuthorship W3024659761A5034476404 @default.
• W3024659761 hasAuthorship W3024659761A5035661649 @default.
• W3024659761 hasAuthorship W3024659761A5037109470 @default.
• W3024659761 hasAuthorship W3024659761A5038895203 @default.
• W3024659761 hasAuthorship W3024659761A5056567731 @default.
• W3024659761 hasAuthorship W3024659761A5057987365 @default.
• W3024659761 hasAuthorship W3024659761A5072495935 @default.
• W3024659761 hasAuthorship W3024659761A5075183307 @default.
• W3024659761 hasBestOaLocation W30246597611 @default.
• W3024659761 hasConcept C127162648 @default.
• W3024659761 hasConcept C153180895 @default.
• W3024659761 hasConcept C154945302 @default.
• W3024659761 hasConcept C2778263558 @default.
• W3024659761 hasConcept C28490314 @default.
• W3024659761 hasConcept C31258907 @default.
• W3024659761 hasConcept C40969351 @default.
• W3024659761 hasConcept C41008148 @default.
• W3024659761 hasConcept C68115822 @default.
• W3024659761 hasConcept C76155785 @default.
• W3024659761 hasConceptScore W3024659761C127162648 @default.
• W3024659761 hasConceptScore W3024659761C153180895 @default.
• W3024659761 hasConceptScore W3024659761C154945302 @default.
• W3024659761 hasConceptScore W3024659761C2778263558 @default.
• W3024659761 hasConceptScore W3024659761C28490314 @default.
• W3024659761 hasConceptScore W3024659761C31258907 @default.
• W3024659761 hasConceptScore W3024659761C40969351 @default.
• W3024659761 hasConceptScore W3024659761C41008148 @default.
• W3024659761 hasConceptScore W3024659761C68115822 @default.
• W3024659761 hasConceptScore W3024659761C76155785 @default.
• W3024659761 hasLocation W30246597611 @default.
• W3024659761 hasOpenAccess W3024659761 @default.
• W3024659761 hasPrimaryLocation W30246597611 @default.
• W3024659761 hasRelatedWork W1544392014 @default.
• W3024659761 hasRelatedWork W1769849273 @default.
• W3024659761 hasRelatedWork W2053255073 @default.
• W3024659761 hasRelatedWork W2280144960 @default.
• W3024659761 hasRelatedWork W2405439032 @default.

• next