Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891609809> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W2891609809 abstract "This paper addresses the problem of recognizing speech corrupted by overlapping speakers in a multichannel setting. To extract a target speaker from the mixture, we use a neural network based beamformer which uses masks estimated by a neural network to compute statistically optimal spatial filters. Following our previous work, we inform the neural network about the target speaker using information extracted from an adaptation utterance’ enabling the network to track the target speaker. While in the previous work, this method was used to separately extract the speaker and then pass such preprocessed speech to a speech recognition system, here we explore training both systems jointly with a common speech recognition criterion. We show that integrating the two systems and training for the final objective improves the performance. In addition, the integration enables further sharing of information between the acoustic model and the speaker extraction system, by making use of the predicted HMM-state posteriors to refine the masks used for beamforming." @default.
- W2891609809 created "2018-09-27" @default.
- W2891609809 creator A5021240106 @default.
- W2891609809 creator A5023868166 @default.
- W2891609809 creator A5045539248 @default.
- W2891609809 creator A5054069029 @default.
- W2891609809 creator A5067790073 @default.
- W2891609809 creator A5069398831 @default.
- W2891609809 date "2018-04-01" @default.
- W2891609809 modified "2023-09-23" @default.
- W2891609809 title "Optimization of Speaker-Aware Multichannel Speech Extraction with ASR Criterion" @default.
- W2891609809 cites W1482149378 @default.
- W2891609809 cites W1987906574 @default.
- W2891609809 cites W2031583051 @default.
- W2891609809 cites W2043216213 @default.
- W2891609809 cites W2129171989 @default.
- W2891609809 cites W2130361043 @default.
- W2891609809 cites W2144763279 @default.
- W2891609809 cites W2158143227 @default.
- W2891609809 cites W2221409856 @default.
- W2891609809 cites W2289394825 @default.
- W2891609809 cites W2397725648 @default.
- W2891609809 cites W2398042854 @default.
- W2891609809 cites W2398776621 @default.
- W2891609809 cites W2398972335 @default.
- W2891609809 cites W2460742184 @default.
- W2891609809 cites W2517616541 @default.
- W2891609809 cites W2589857635 @default.
- W2891609809 cites W2591810467 @default.
- W2891609809 cites W2640112133 @default.
- W2891609809 cites W2699350432 @default.
- W2891609809 cites W2714487941 @default.
- W2891609809 cites W2749510669 @default.
- W2891609809 cites W2750446090 @default.
- W2891609809 cites W4205778870 @default.
- W2891609809 doi "https://doi.org/10.1109/icassp.2018.8461533" @default.
- W2891609809 hasPublicationYear "2018" @default.
- W2891609809 type Work @default.
- W2891609809 sameAs 2891609809 @default.
- W2891609809 citedByCount "6" @default.
- W2891609809 countsByYear W28916098092018 @default.
- W2891609809 countsByYear W28916098092019 @default.
- W2891609809 countsByYear W28916098092022 @default.
- W2891609809 crossrefType "proceedings-article" @default.
- W2891609809 hasAuthorship W2891609809A5021240106 @default.
- W2891609809 hasAuthorship W2891609809A5023868166 @default.
- W2891609809 hasAuthorship W2891609809A5045539248 @default.
- W2891609809 hasAuthorship W2891609809A5054069029 @default.
- W2891609809 hasAuthorship W2891609809A5067790073 @default.
- W2891609809 hasAuthorship W2891609809A5069398831 @default.
- W2891609809 hasConcept C133892786 @default.
- W2891609809 hasConcept C154945302 @default.
- W2891609809 hasConcept C28490314 @default.
- W2891609809 hasConcept C41008148 @default.
- W2891609809 hasConceptScore W2891609809C133892786 @default.
- W2891609809 hasConceptScore W2891609809C154945302 @default.
- W2891609809 hasConceptScore W2891609809C28490314 @default.
- W2891609809 hasConceptScore W2891609809C41008148 @default.
- W2891609809 hasLocation W28916098091 @default.
- W2891609809 hasOpenAccess W2891609809 @default.
- W2891609809 hasPrimaryLocation W28916098091 @default.
- W2891609809 hasRelatedWork W1491159402 @default.
- W2891609809 hasRelatedWork W2312116756 @default.
- W2891609809 hasRelatedWork W2368779261 @default.
- W2891609809 hasRelatedWork W2499802997 @default.
- W2891609809 hasRelatedWork W2778699561 @default.
- W2891609809 hasRelatedWork W2794438528 @default.
- W2891609809 hasRelatedWork W2893763841 @default.
- W2891609809 hasRelatedWork W3162054169 @default.
- W2891609809 hasRelatedWork W4297807400 @default.
- W2891609809 hasRelatedWork W2175373321 @default.
- W2891609809 isParatext "false" @default.
- W2891609809 isRetracted "false" @default.
- W2891609809 magId "2891609809" @default.
- W2891609809 workType "article" @default.