FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2099881084> ?p ?o ?g. }

• W2099881084 endingPage "1113" @default.
• W2099881084 startingPage "1098" @default.
• W2099881084 abstract "In this paper, a set of feature vector normalization methods based on the minimum mean square error (MMSE) criterion and stereo data is presented. They include multi-environment model-based linear normalization (MEMLIN), polynomial MEMLIN (P-MEMLIN), multi-environment model-based histogram normalization (MEMHIN), and phoneme-dependent MEMLIN (PD-MEMLIN). Those methods model clean and noisy feature vector spaces using Gaussian mixture models (GMMs). The objective of the methods is to learn a transformation between clean and noisy feature vectors associated with each pair of clean and noisy model Gaussians. The direct approach to learn the transformation is by using stereo data; that is, noisy feature vectors and the corresponding clean feature vectors. In this paper, however, a nonstereo data based training procedure, is presented. The transformations can be modeled just like a bias vector (MEMLIN), or by using a first-order polynomial (P-MEMLIN) or a nonlinear function based on histogram equalization (MEMHIN). Further improvements are obtained by using phoneme-dependent bias vector transformation (PD-MEMLIN). In PD-MEMLIN, the clean and noisy feature vector spaces are split into several phonemes, and each of them is modeled as a GMM. Those methods achieve significant word error rate improvements over others that are based on similar targets. The experimental results using the SpeechDat Car database show an average improvement in word error rate greater than 68% in all cases compared to the baseline when using the original clean acoustic models, and up to 83% when training acoustic models on the new normalized feature space" @default.
• W2099881084 created "2016-06-24" @default.
• W2099881084 creator A5020174874 @default.
• W2099881084 creator A5036493563 @default.
• W2099881084 creator A5058841753 @default.
• W2099881084 creator A5066085685 @default.
• W2099881084 creator A5086487784 @default.
• W2099881084 date "2007-03-01" @default.
• W2099881084 modified "2023-10-01" @default.
• W2099881084 title "Cepstral Vector Normalization Based on Stereo Data for Robust Speech Recognition" @default.
• W2099881084 cites W1279245896 @default.
• W2099881084 cites W1665196592 @default.
• W2099881084 cites W168238592 @default.
• W2099881084 cites W177521823 @default.
• W2099881084 cites W183860 @default.
• W2099881084 cites W190920181 @default.
• W2099881084 cites W1965555277 @default.
• W2099881084 cites W2036099872 @default.
• W2099881084 cites W2082474452 @default.
• W2099881084 cites W2100969003 @default.
• W2099881084 cites W2108296887 @default.
• W2099881084 cites W2113703812 @default.
• W2099881084 cites W2113911479 @default.
• W2099881084 cites W2121973264 @default.
• W2099881084 cites W2128653836 @default.
• W2099881084 cites W2129967803 @default.
• W2099881084 cites W2137075158 @default.
• W2099881084 cites W2137418074 @default.
• W2099881084 cites W2146871184 @default.
• W2099881084 cites W2148357360 @default.
• W2099881084 cites W2157590573 @default.
• W2099881084 cites W2157702502 @default.
• W2099881084 cites W2407641997 @default.
• W2099881084 cites W68550219 @default.
• W2099881084 doi "https://doi.org/10.1109/tasl.2006.885244" @default.
• W2099881084 hasPublicationYear "2007" @default.
• W2099881084 type Work @default.
• W2099881084 sameAs 2099881084 @default.
• W2099881084 citedByCount "33" @default.
• W2099881084 countsByYear W20998810842012 @default.
• W2099881084 countsByYear W20998810842013 @default.
• W2099881084 countsByYear W20998810842014 @default.
• W2099881084 countsByYear W20998810842015 @default.
• W2099881084 countsByYear W20998810842017 @default.
• W2099881084 countsByYear W20998810842019 @default.
• W2099881084 countsByYear W20998810842020 @default.
• W2099881084 countsByYear W20998810842022 @default.
• W2099881084 countsByYear W20998810842023 @default.
• W2099881084 crossrefType "journal-article" @default.
• W2099881084 hasAuthorship W2099881084A5020174874 @default.
• W2099881084 hasAuthorship W2099881084A5036493563 @default.
• W2099881084 hasAuthorship W2099881084A5058841753 @default.
• W2099881084 hasAuthorship W2099881084A5066085685 @default.
• W2099881084 hasAuthorship W2099881084A5086487784 @default.
• W2099881084 hasConcept C115961682 @default.
• W2099881084 hasConcept C136886441 @default.
• W2099881084 hasConcept C138885662 @default.
• W2099881084 hasConcept C144024400 @default.
• W2099881084 hasConcept C153180895 @default.
• W2099881084 hasConcept C154945302 @default.
• W2099881084 hasConcept C19165224 @default.
• W2099881084 hasConcept C2776401178 @default.
• W2099881084 hasConcept C28490314 @default.
• W2099881084 hasConcept C33923547 @default.
• W2099881084 hasConcept C40969351 @default.
• W2099881084 hasConcept C41008148 @default.
• W2099881084 hasConcept C41895202 @default.
• W2099881084 hasConcept C53533937 @default.
• W2099881084 hasConcept C61224824 @default.
• W2099881084 hasConcept C83665646 @default.
• W2099881084 hasConceptScore W2099881084C115961682 @default.
• W2099881084 hasConceptScore W2099881084C136886441 @default.
• W2099881084 hasConceptScore W2099881084C138885662 @default.
• W2099881084 hasConceptScore W2099881084C144024400 @default.
• W2099881084 hasConceptScore W2099881084C153180895 @default.
• W2099881084 hasConceptScore W2099881084C154945302 @default.
• W2099881084 hasConceptScore W2099881084C19165224 @default.
• W2099881084 hasConceptScore W2099881084C2776401178 @default.
• W2099881084 hasConceptScore W2099881084C28490314 @default.
• W2099881084 hasConceptScore W2099881084C33923547 @default.
• W2099881084 hasConceptScore W2099881084C40969351 @default.
• W2099881084 hasConceptScore W2099881084C41008148 @default.
• W2099881084 hasConceptScore W2099881084C41895202 @default.
• W2099881084 hasConceptScore W2099881084C53533937 @default.
• W2099881084 hasConceptScore W2099881084C61224824 @default.
• W2099881084 hasConceptScore W2099881084C83665646 @default.
• W2099881084 hasIssue "3" @default.
• W2099881084 hasLocation W20998810841 @default.
• W2099881084 hasOpenAccess W2099881084 @default.
• W2099881084 hasPrimaryLocation W20998810841 @default.
• W2099881084 hasRelatedWork W1988928367 @default.
• W2099881084 hasRelatedWork W1991269640 @default.
• W2099881084 hasRelatedWork W2052112670 @default.
• W2099881084 hasRelatedWork W2052253960 @default.
• W2099881084 hasRelatedWork W2066259560 @default.
• W2099881084 hasRelatedWork W2085466496 @default.
• W2099881084 hasRelatedWork W2098144668 @default.
• W2099881084 hasRelatedWork W2114072031 @default.

• next