FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W280592816> ?p ?o ?g. }

• W280592816 abstract "This paper describes an efficient online target-speech-extraction method used as a preprocessing step for robust automatic speech recognition (ASR). Because a target speaker is located relatively close to microphones in many ASR applications, acoustic paths to microphones are moderately reverberant, and the target speaker direction can easily be estimated. In this situation, noise estimation is effectively performed by forming a directional null to the target speaker. Required weights for extracting target speech, independent of the estimated noise, are then determined using an adaptation rule derived from a modified version of the cost function for independent component analysis (ICA), while retaining the minimal distortion principle. In particular, an online natural-gradient learning rule with a nonholonomic constraint and normalization by a smoothed power estimate of the input signal is derived for stable convergence, even for dynamically changing speech levels, with much less computational complexity than conventional ICA. Furthermore, stereo mixtures are considered as input data for further reduction of computational loads and fast convergence. Although the method may suffer from the underdetermined problem, the weights are adapted to obtain signal-to-noise-ratio-maximization beamformers for successful target speech estimation. The experimental results obtained for various conditions demonstrate the effectiveness of the proposed method. HighlightsAn efficient online preprocessing method is proposed for robust speech recognition.The method extracts target speech with target speaker direction as a prior knowledge.A parameter adaptation rule was derived by modifying independent component analysis.The method achieves robustness by adapting parameters with fixed noise estimation.On average, the method showed better recognition performance with fewer computations." @default.
• W280592816 created "2016-06-24" @default.
• W280592816 creator A5024926218 @default.
• W280592816 creator A5031280750 @default.
• W280592816 date "2015-12-01" @default.
• W280592816 modified "2023-09-22" @default.
• W280592816 title "Efficient online target speech extraction using DOA-constrained independent component analysis of stereo data for robust speech recognition" @default.
• W280592816 cites W102482883 @default.
• W280592816 cites W1510095622 @default.
• W280592816 cites W1580676799 @default.
• W280592816 cites W1643320849 @default.
• W280592816 cites W1755563775 @default.
• W280592816 cites W18621165 @default.
• W280592816 cites W1979800130 @default.
• W280592816 cites W1986159300 @default.
• W280592816 cites W2006547268 @default.
• W280592816 cites W2027884847 @default.
• W280592816 cites W2046239551 @default.
• W280592816 cites W2062572859 @default.
• W280592816 cites W2075633727 @default.
• W280592816 cites W2096855653 @default.
• W280592816 cites W2098723113 @default.
• W280592816 cites W2105617050 @default.
• W280592816 cites W2108357193 @default.
• W280592816 cites W2113198618 @default.
• W280592816 cites W2114461480 @default.
• W280592816 cites W2117678320 @default.
• W280592816 cites W2118078048 @default.
• W280592816 cites W2128911505 @default.
• W280592816 cites W2134910060 @default.
• W280592816 cites W2151484683 @default.
• W280592816 cites W2159485234 @default.
• W280592816 cites W2161270956 @default.
• W280592816 cites W2164162335 @default.
• W280592816 cites W2170768669 @default.
• W280592816 cites W2334833135 @default.
• W280592816 cites W2435167693 @default.
• W280592816 cites W9420449 @default.
• W280592816 cites W2342337667 @default.
• W280592816 cites W3031363333 @default.
• W280592816 doi "https://doi.org/10.1016/j.sigpro.2015.04.022" @default.
• W280592816 hasPublicationYear "2015" @default.
• W280592816 type Work @default.
• W280592816 sameAs 280592816 @default.
• W280592816 citedByCount "9" @default.
• W280592816 countsByYear W2805928162016 @default.
• W280592816 countsByYear W2805928162017 @default.
• W280592816 countsByYear W2805928162018 @default.
• W280592816 countsByYear W2805928162019 @default.
• W280592816 countsByYear W2805928162020 @default.
• W280592816 crossrefType "journal-article" @default.
• W280592816 hasAuthorship W280592816A5024926218 @default.
• W280592816 hasAuthorship W280592816A5031280750 @default.
• W280592816 hasConcept C104317684 @default.
• W280592816 hasConcept C153180895 @default.
• W280592816 hasConcept C154945302 @default.
• W280592816 hasConcept C163294075 @default.
• W280592816 hasConcept C185592680 @default.
• W280592816 hasConcept C2776182073 @default.
• W280592816 hasConcept C28490314 @default.
• W280592816 hasConcept C34736171 @default.
• W280592816 hasConcept C41008148 @default.
• W280592816 hasConcept C51432778 @default.
• W280592816 hasConcept C55493867 @default.
• W280592816 hasConcept C61328038 @default.
• W280592816 hasConcept C63479239 @default.
• W280592816 hasConceptScore W280592816C104317684 @default.
• W280592816 hasConceptScore W280592816C153180895 @default.
• W280592816 hasConceptScore W280592816C154945302 @default.
• W280592816 hasConceptScore W280592816C163294075 @default.
• W280592816 hasConceptScore W280592816C185592680 @default.
• W280592816 hasConceptScore W280592816C2776182073 @default.
• W280592816 hasConceptScore W280592816C28490314 @default.
• W280592816 hasConceptScore W280592816C34736171 @default.
• W280592816 hasConceptScore W280592816C41008148 @default.
• W280592816 hasConceptScore W280592816C51432778 @default.
• W280592816 hasConceptScore W280592816C55493867 @default.
• W280592816 hasConceptScore W280592816C61328038 @default.
• W280592816 hasConceptScore W280592816C63479239 @default.
• W280592816 hasFunder F4320322030 @default.
• W280592816 hasLocation W2805928161 @default.
• W280592816 hasOpenAccess W280592816 @default.
• W280592816 hasPrimaryLocation W2805928161 @default.
• W280592816 hasRelatedWork W1664684929 @default.
• W280592816 hasRelatedWork W1932090834 @default.
• W280592816 hasRelatedWork W1995828869 @default.
• W280592816 hasRelatedWork W2004838643 @default.
• W280592816 hasRelatedWork W2007589241 @default.
• W280592816 hasRelatedWork W204325298 @default.
• W280592816 hasRelatedWork W2053371020 @default.
• W280592816 hasRelatedWork W2065080272 @default.
• W280592816 hasRelatedWork W2076979746 @default.
• W280592816 hasRelatedWork W2113911479 @default.
• W280592816 hasRelatedWork W2117857642 @default.
• W280592816 hasRelatedWork W2121388655 @default.
• W280592816 hasRelatedWork W2140595311 @default.
• W280592816 hasRelatedWork W2162130745 @default.
• W280592816 hasRelatedWork W2167204516 @default.
• W280592816 hasRelatedWork W2285729352 @default.
• W280592816 hasRelatedWork W2564013664 @default.

• next