FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2560544917> ?p ?o ?g. }

• W2560544917 abstract "Abstract In this paper we propose to exploit the automatic Quality Estimation (QE) of ASR hypotheses to perform the unsupervised adaptation of a deep neural network modeling acoustic probabilities. Our hypothesis is that significant improvements can be achieved by: i) automatically transcribing the evaluation data we are currently trying to recognise, and ii) selecting from it a subset of “good quality” instances based on the word error rate (WER) scores predicted by a QE component. To validate this hypothesis, we run several experiments on the evaluation data sets released for the CHiME-3 challenge. First, we operate in oracle conditions in which manual transcriptions of the evaluation data are available, thus allowing us to compute the true sentence WER. In this scenario, we perform the adaptation with variable amounts of data, which are characterised by different levels of quality. Then, we move to realistic conditions in which the manual transcriptions of the evaluation data are not available. In this case, the adaptation is performed on data selected according to the WER scores predicted by a QE component. Our results indicate that: i) QE predictions allow us to closely approximate the adaptation results obtained in oracle conditions, and ii) the overall ASR performance based on the proposed QE-driven adaptation method is significantly better than the strong, most recent, CHiME-3 baseline." @default.
• W2560544917 created "2016-12-16" @default.
• W2560544917 creator A5015363344 @default.
• W2560544917 creator A5025827839 @default.
• W2560544917 creator A5037923499 @default.
• W2560544917 creator A5041102462 @default.
• W2560544917 creator A5091420288 @default.
• W2560544917 date "2017-11-01" @default.
• W2560544917 modified "2023-09-24" @default.
• W2560544917 title "DNN adaptation by automatic quality estimation of ASR hypotheses" @default.
• W2560544917 cites W1513862252 @default.
• W2560544917 cites W1524333225 @default.
• W2560544917 cites W1537275613 @default.
• W2560544917 cites W1596940414 @default.
• W2560544917 cites W1654858490 @default.
• W2560544917 cites W1904457459 @default.
• W2560544917 cites W1981617403 @default.
• W2560544917 cites W1989549063 @default.
• W2560544917 cites W1993660824 @default.
• W2560544917 cites W1993882792 @default.
• W2560544917 cites W2002342963 @default.
• W2560544917 cites W2036351241 @default.
• W2560544917 cites W2047817868 @default.
• W2560544917 cites W2048526313 @default.
• W2560544917 cites W2056132907 @default.
• W2560544917 cites W2062164080 @default.
• W2560544917 cites W2070707809 @default.
• W2560544917 cites W2080005694 @default.
• W2560544917 cites W2087006792 @default.
• W2560544917 cites W2094147890 @default.
• W2560544917 cites W2101045344 @default.
• W2560544917 cites W2102113734 @default.
• W2560544917 cites W2129334286 @default.
• W2560544917 cites W2130037349 @default.
• W2560544917 cites W2131342762 @default.
• W2560544917 cites W2136879537 @default.
• W2560544917 cites W2143860348 @default.
• W2560544917 cites W2147768505 @default.
• W2560544917 cites W2148613904 @default.
• W2560544917 cites W2149791639 @default.
• W2560544917 cites W2160306971 @default.
• W2560544917 cites W2160815625 @default.
• W2560544917 cites W2161482971 @default.
• W2560544917 cites W2162042984 @default.
• W2560544917 cites W2169939314 @default.
• W2560544917 cites W2184045248 @default.
• W2560544917 cites W2187166234 @default.
• W2560544917 cites W2193413348 @default.
• W2560544917 cites W2250799045 @default.
• W2560544917 cites W2251880053 @default.
• W2560544917 cites W2289394825 @default.
• W2560544917 cites W2289731793 @default.
• W2560544917 cites W2290469434 @default.
• W2560544917 cites W2294161614 @default.
• W2560544917 cites W2394932179 @default.
• W2560544917 cites W2396230943 @default.
• W2560544917 cites W2396422584 @default.
• W2560544917 cites W2404620314 @default.
• W2560544917 cites W2517015344 @default.
• W2560544917 cites W2594610113 @default.
• W2560544917 cites W2616139854 @default.
• W2560544917 cites W61355150 @default.
• W2560544917 cites W651894412 @default.
• W2560544917 cites W82936479 @default.
• W2560544917 cites W2555431843 @default.
• W2560544917 doi "https://doi.org/10.1016/j.csl.2016.11.002" @default.
• W2560544917 hasPublicationYear "2017" @default.
• W2560544917 type Work @default.
• W2560544917 sameAs 2560544917 @default.
• W2560544917 citedByCount "5" @default.
• W2560544917 countsByYear W25605449172017 @default.
• W2560544917 countsByYear W25605449172018 @default.
• W2560544917 countsByYear W25605449172019 @default.
• W2560544917 crossrefType "journal-article" @default.
• W2560544917 hasAuthorship W2560544917A5015363344 @default.
• W2560544917 hasAuthorship W2560544917A5025827839 @default.
• W2560544917 hasAuthorship W2560544917A5037923499 @default.
• W2560544917 hasAuthorship W2560544917A5041102462 @default.
• W2560544917 hasAuthorship W2560544917A5091420288 @default.
• W2560544917 hasBestOaLocation W25605449172 @default.
• W2560544917 hasConcept C111368507 @default.
• W2560544917 hasConcept C111472728 @default.
• W2560544917 hasConcept C115903868 @default.
• W2560544917 hasConcept C120665830 @default.
• W2560544917 hasConcept C121332964 @default.
• W2560544917 hasConcept C12725497 @default.
• W2560544917 hasConcept C127313418 @default.
• W2560544917 hasConcept C138885662 @default.
• W2560544917 hasConcept C139807058 @default.
• W2560544917 hasConcept C154945302 @default.
• W2560544917 hasConcept C165696696 @default.
• W2560544917 hasConcept C2524010 @default.
• W2560544917 hasConcept C2777530160 @default.
• W2560544917 hasConcept C2779530757 @default.
• W2560544917 hasConcept C28490314 @default.
• W2560544917 hasConcept C2984842247 @default.
• W2560544917 hasConcept C33923547 @default.
• W2560544917 hasConcept C38652104 @default.
• W2560544917 hasConcept C40969351 @default.
• W2560544917 hasConcept C41008148 @default.

• next