Matches in SemOpenAlex for { <https://semopenalex.org/work/W2772028198> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W2772028198 abstract "Lattice generation is an essential feature of the decoder for many speech recognition applications. In this paper, we first review lattice generation methods for WFST-based decoding and describe in a uniform formalism two established approaches for state-of-the-art speech recognition systems: the phone pair and the N-best histories approaches. We then present a novel optimization method, pruned determinization followed by minimization, that produces a deterministic minimal lattice that retains all paths within specified weight and lattice size thresholds. Experimentally, we show that before optimization, the phone-pair and the N-best histories approaches each have conditions where they perform better when evaluated on video transcription and mixed voice search and dictation tasks. However, once this lattice optimization procedure is applied, the phone pair approach has the lowest oracle WER for a given lattice density by a significant margin. We further show that the pruned determinization presented here is efficient to use during decoding unlike classical weighted determinization from which it is derived. Finally, we consider on-the-fly lattice rescoring in which the lattice generation and combination with the secondary LM are done in one step. We compare the phone pair and N-best histories approaches for this scenario and find the former superior in our experiments." @default.
- W2772028198 created "2017-12-22" @default.
- W2772028198 creator A5014137885 @default.
- W2772028198 creator A5050133412 @default.
- W2772028198 creator A5081836495 @default.
- W2772028198 date "2017-12-01" @default.
- W2772028198 modified "2023-09-25" @default.
- W2772028198 title "On lattice generation for large vocabulary speech recognition" @default.
- W2772028198 cites W143425341 @default.
- W2772028198 cites W1524975733 @default.
- W2772028198 cites W1582482241 @default.
- W2772028198 cites W1629199664 @default.
- W2772028198 cites W173514433 @default.
- W2772028198 cites W1884028435 @default.
- W2772028198 cites W1895481600 @default.
- W2772028198 cites W1969483458 @default.
- W2772028198 cites W1977697308 @default.
- W2772028198 cites W200927961 @default.
- W2772028198 cites W2022951240 @default.
- W2772028198 cites W205749991 @default.
- W2772028198 cites W2098318492 @default.
- W2772028198 cites W2116047093 @default.
- W2772028198 cites W2124912321 @default.
- W2772028198 cites W2127141656 @default.
- W2772028198 cites W2132991150 @default.
- W2772028198 cites W2137446469 @default.
- W2772028198 cites W2138889249 @default.
- W2772028198 cites W2143919998 @default.
- W2772028198 cites W2157953146 @default.
- W2772028198 cites W2163766511 @default.
- W2772028198 cites W2164579587 @default.
- W2772028198 cites W2164714022 @default.
- W2772028198 cites W2293829681 @default.
- W2772028198 cites W2403939550 @default.
- W2772028198 cites W7198584 @default.
- W2772028198 doi "https://doi.org/10.1109/asru.2017.8268940" @default.
- W2772028198 hasPublicationYear "2017" @default.
- W2772028198 type Work @default.
- W2772028198 sameAs 2772028198 @default.
- W2772028198 citedByCount "5" @default.
- W2772028198 countsByYear W27720281982019 @default.
- W2772028198 countsByYear W27720281982021 @default.
- W2772028198 countsByYear W27720281982022 @default.
- W2772028198 crossrefType "proceedings-article" @default.
- W2772028198 hasAuthorship W2772028198A5014137885 @default.
- W2772028198 hasAuthorship W2772028198A5050133412 @default.
- W2772028198 hasAuthorship W2772028198A5081836495 @default.
- W2772028198 hasConcept C11413529 @default.
- W2772028198 hasConcept C115903868 @default.
- W2772028198 hasConcept C121332964 @default.
- W2772028198 hasConcept C138885662 @default.
- W2772028198 hasConcept C154945302 @default.
- W2772028198 hasConcept C24890656 @default.
- W2772028198 hasConcept C2778707766 @default.
- W2772028198 hasConcept C2781204021 @default.
- W2772028198 hasConcept C28490314 @default.
- W2772028198 hasConcept C41008148 @default.
- W2772028198 hasConcept C41895202 @default.
- W2772028198 hasConcept C55166926 @default.
- W2772028198 hasConcept C57273362 @default.
- W2772028198 hasConcept C95623464 @default.
- W2772028198 hasConceptScore W2772028198C11413529 @default.
- W2772028198 hasConceptScore W2772028198C115903868 @default.
- W2772028198 hasConceptScore W2772028198C121332964 @default.
- W2772028198 hasConceptScore W2772028198C138885662 @default.
- W2772028198 hasConceptScore W2772028198C154945302 @default.
- W2772028198 hasConceptScore W2772028198C24890656 @default.
- W2772028198 hasConceptScore W2772028198C2778707766 @default.
- W2772028198 hasConceptScore W2772028198C2781204021 @default.
- W2772028198 hasConceptScore W2772028198C28490314 @default.
- W2772028198 hasConceptScore W2772028198C41008148 @default.
- W2772028198 hasConceptScore W2772028198C41895202 @default.
- W2772028198 hasConceptScore W2772028198C55166926 @default.
- W2772028198 hasConceptScore W2772028198C57273362 @default.
- W2772028198 hasConceptScore W2772028198C95623464 @default.
- W2772028198 hasLocation W27720281981 @default.
- W2772028198 hasOpenAccess W2772028198 @default.
- W2772028198 hasPrimaryLocation W27720281981 @default.
- W2772028198 hasRelatedWork W11531451 @default.
- W2772028198 hasRelatedWork W1552159754 @default.
- W2772028198 hasRelatedWork W2148757832 @default.
- W2772028198 hasRelatedWork W2151447942 @default.
- W2772028198 hasRelatedWork W2293457016 @default.
- W2772028198 hasRelatedWork W2368651715 @default.
- W2772028198 hasRelatedWork W2611614995 @default.
- W2772028198 hasRelatedWork W2789919619 @default.
- W2772028198 hasRelatedWork W2900389025 @default.
- W2772028198 hasRelatedWork W3107474891 @default.
- W2772028198 isParatext "false" @default.
- W2772028198 isRetracted "false" @default.
- W2772028198 magId "2772028198" @default.
- W2772028198 workType "article" @default.