FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213303412> ?p ?o ?g. }

• W3213303412 endingPage "104541" @default.
• W3213303412 startingPage "104541" @default.
• W3213303412 abstract "To detect social signals such as laughter and filler events in an audio recording, the most straightforward way is to utilize a Hidden Markov Model — or these days a Hidden Markov Model/Deep Neural Network (HMM/DNN) hybrid. HMM/DNNs, however, perform best if the DNN outputs are scaled by dividing them by the a priori class probabilities first, before applying a dynamic or Viterbi beam search. These class a priori probability values (or priors for short) are usually estimated by counting the frame occurrences of each class in the training set and then dividing these totals by the total number of frames. These estimates, however, may in fact be suboptimal for a number of reasons ranging from imprecise labeling to the overconfidence of DNNs. In this study we show empirically that more reliable scaling factors can be obtained by optimization. Using this approach, we managed to achieve a 6−9% relative error reduction both at the frame level and the segment level, using a public database containing spontaneous English mobile phone conversations." @default.
• W3213303412 created "2021-11-22" @default.
• W3213303412 creator A5088559776 @default.
• W3213303412 date "2022-01-01" @default.
• W3213303412 modified "2023-09-23" @default.
• W3213303412 title "Optimizing class priors to improve the detection of social signals in audio data" @default.
• W3213303412 cites W1605438009 @default.
• W3213303412 cites W1656711680 @default.
• W3213303412 cites W1919365417 @default.
• W3213303412 cites W1972047492 @default.
• W3213303412 cites W1972978214 @default.
• W3213303412 cites W1976349544 @default.
• W3213303412 cites W1976697056 @default.
• W3213303412 cites W1993232824 @default.
• W3213303412 cites W1993882792 @default.
• W3213303412 cites W2004227461 @default.
• W3213303412 cites W2005199108 @default.
• W3213303412 cites W2012787434 @default.
• W3213303412 cites W2061436640 @default.
• W3213303412 cites W2064675550 @default.
• W3213303412 cites W2085662862 @default.
• W3213303412 cites W2111084382 @default.
• W3213303412 cites W2112036188 @default.
• W3213303412 cites W2134473538 @default.
• W3213303412 cites W2141253686 @default.
• W3213303412 cites W2144005487 @default.
• W3213303412 cites W2157331557 @default.
• W3213303412 cites W2171060319 @default.
• W3213303412 cites W2286192402 @default.
• W3213303412 cites W2403021840 @default.
• W3213303412 cites W2404132853 @default.
• W3213303412 cites W2482033662 @default.
• W3213303412 cites W2518276686 @default.
• W3213303412 cites W2565345033 @default.
• W3213303412 cites W2566895423 @default.
• W3213303412 cites W2747786714 @default.
• W3213303412 cites W2767106145 @default.
• W3213303412 cites W2887590219 @default.
• W3213303412 cites W2890222510 @default.
• W3213303412 cites W2895625244 @default.
• W3213303412 cites W2903782945 @default.
• W3213303412 cites W2904408089 @default.
• W3213303412 cites W2936503027 @default.
• W3213303412 cites W2965526224 @default.
• W3213303412 cites W2968379855 @default.
• W3213303412 cites W2972944488 @default.
• W3213303412 cites W2998994232 @default.
• W3213303412 cites W3099955943 @default.
• W3213303412 cites W3123899751 @default.
• W3213303412 doi "https://doi.org/10.1016/j.engappai.2021.104541" @default.
• W3213303412 hasPublicationYear "2022" @default.
• W3213303412 type Work @default.
• W3213303412 sameAs 3213303412 @default.
• W3213303412 citedByCount "1" @default.
• W3213303412 countsByYear W32133034122022 @default.
• W3213303412 crossrefType "journal-article" @default.
• W3213303412 hasAuthorship W3213303412A5088559776 @default.
• W3213303412 hasConcept C107673813 @default.
• W3213303412 hasConcept C111472728 @default.
• W3213303412 hasConcept C126042441 @default.
• W3213303412 hasConcept C138885662 @default.
• W3213303412 hasConcept C153180895 @default.
• W3213303412 hasConcept C154945302 @default.
• W3213303412 hasConcept C177264268 @default.
• W3213303412 hasConcept C177769412 @default.
• W3213303412 hasConcept C199360897 @default.
• W3213303412 hasConcept C23224414 @default.
• W3213303412 hasConcept C2777212361 @default.
• W3213303412 hasConcept C28490314 @default.
• W3213303412 hasConcept C41008148 @default.
• W3213303412 hasConcept C60582962 @default.
• W3213303412 hasConcept C75553542 @default.
• W3213303412 hasConcept C76155785 @default.
• W3213303412 hasConceptScore W3213303412C107673813 @default.
• W3213303412 hasConceptScore W3213303412C111472728 @default.
• W3213303412 hasConceptScore W3213303412C126042441 @default.
• W3213303412 hasConceptScore W3213303412C138885662 @default.
• W3213303412 hasConceptScore W3213303412C153180895 @default.
• W3213303412 hasConceptScore W3213303412C154945302 @default.
• W3213303412 hasConceptScore W3213303412C177264268 @default.
• W3213303412 hasConceptScore W3213303412C177769412 @default.
• W3213303412 hasConceptScore W3213303412C199360897 @default.
• W3213303412 hasConceptScore W3213303412C23224414 @default.
• W3213303412 hasConceptScore W3213303412C2777212361 @default.
• W3213303412 hasConceptScore W3213303412C28490314 @default.
• W3213303412 hasConceptScore W3213303412C41008148 @default.
• W3213303412 hasConceptScore W3213303412C60582962 @default.
• W3213303412 hasConceptScore W3213303412C75553542 @default.
• W3213303412 hasConceptScore W3213303412C76155785 @default.
• W3213303412 hasLocation W32133034121 @default.
• W3213303412 hasOpenAccess W3213303412 @default.
• W3213303412 hasPrimaryLocation W32133034121 @default.
• W3213303412 hasRelatedWork W1505908888 @default.
• W3213303412 hasRelatedWork W1554908322 @default.
• W3213303412 hasRelatedWork W1896748021 @default.
• W3213303412 hasRelatedWork W2019307865 @default.
• W3213303412 hasRelatedWork W2100982643 @default.
• W3213303412 hasRelatedWork W2107432317 @default.

• next