Matches in SemOpenAlex for { <https://semopenalex.org/work/W2024504639> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W2024504639 endingPage "P02004" @default.
- W2024504639 startingPage "P02004" @default.
- W2024504639 abstract "Speech processing has benefited a great deal from wavelet transforms. Wavelet packets decompose signals into broader components using linear spectral bisecting. The mixing matrix is the key issue in the blind source separation literature especially in underdetermined cases (more sources than sensors). In this paper, algorithms are proposed for estimating the mixing matrix and separation of speech signals from noise free linear mixtures in overcomplete cases. Mixtures of speech signals are decomposed using wavelet packets, the phase difference between the two mixtures is defined and used in the wavelet domain, and histograms of phase differences are obtained for every wavelet packet. In our method, the Laplacian mixture model is considered in the wavelet packet domain, and is applied to each histogram of packets. An expectation maximization algorithm is used to train the model and calculate the model parameters. We also propose a novel method for obtaining the best wavelet packet node for finding source directions in scatter plots using variance calculations. A comparison is made to evaluate the performances of different mother wavelets in the estimation of the mixing matrix and the best wavelet has been chosen. On the basis of the geometrical model, a two-step adaptive algorithm is proposed for separating sources from mixtures." @default.
- W2024504639 created "2016-06-24" @default.
- W2024504639 creator A5008153115 @default.
- W2024504639 creator A5077043805 @default.
- W2024504639 date "2007-02-05" @default.
- W2024504639 modified "2023-09-27" @default.
- W2024504639 title "Blind source separation of speech sources in wavelet packet domains using Laplacian mixture model expectation maximization estimation in overcomplete cases" @default.
- W2024504639 cites W1970352604 @default.
- W2024504639 cites W1978831713 @default.
- W2024504639 cites W1987906574 @default.
- W2024504639 cites W2000631354 @default.
- W2024504639 cites W2020062309 @default.
- W2024504639 cites W2038237443 @default.
- W2024504639 cites W2044581981 @default.
- W2024504639 cites W2061659108 @default.
- W2024504639 cites W2062018842 @default.
- W2024504639 cites W2072570381 @default.
- W2024504639 cites W2099741732 @default.
- W2024504639 cites W2108384452 @default.
- W2024504639 cites W2116799974 @default.
- W2024504639 cites W2119647652 @default.
- W2024504639 cites W2119837299 @default.
- W2024504639 cites W2129451521 @default.
- W2024504639 cites W2132984323 @default.
- W2024504639 cites W2140499889 @default.
- W2024504639 cites W2152910484 @default.
- W2024504639 cites W2166159048 @default.
- W2024504639 cites W4242656336 @default.
- W2024504639 cites W4310638450 @default.
- W2024504639 doi "https://doi.org/10.1088/1742-5468/2007/02/p02004" @default.
- W2024504639 hasPublicationYear "2007" @default.
- W2024504639 type Work @default.
- W2024504639 sameAs 2024504639 @default.
- W2024504639 citedByCount "5" @default.
- W2024504639 countsByYear W20245046392012 @default.
- W2024504639 countsByYear W20245046392017 @default.
- W2024504639 crossrefType "journal-article" @default.
- W2024504639 hasAuthorship W2024504639A5008153115 @default.
- W2024504639 hasAuthorship W2024504639A5077043805 @default.
- W2024504639 hasConcept C105795698 @default.
- W2024504639 hasConcept C119857082 @default.
- W2024504639 hasConcept C120317606 @default.
- W2024504639 hasConcept C127162648 @default.
- W2024504639 hasConcept C153180895 @default.
- W2024504639 hasConcept C154945302 @default.
- W2024504639 hasConcept C155777637 @default.
- W2024504639 hasConcept C182081679 @default.
- W2024504639 hasConcept C196216189 @default.
- W2024504639 hasConcept C2776061190 @default.
- W2024504639 hasConcept C28490314 @default.
- W2024504639 hasConcept C33923547 @default.
- W2024504639 hasConcept C41008148 @default.
- W2024504639 hasConcept C47432892 @default.
- W2024504639 hasConcept C49781872 @default.
- W2024504639 hasConcept C61224824 @default.
- W2024504639 hasConcept C76155785 @default.
- W2024504639 hasConceptScore W2024504639C105795698 @default.
- W2024504639 hasConceptScore W2024504639C119857082 @default.
- W2024504639 hasConceptScore W2024504639C120317606 @default.
- W2024504639 hasConceptScore W2024504639C127162648 @default.
- W2024504639 hasConceptScore W2024504639C153180895 @default.
- W2024504639 hasConceptScore W2024504639C154945302 @default.
- W2024504639 hasConceptScore W2024504639C155777637 @default.
- W2024504639 hasConceptScore W2024504639C182081679 @default.
- W2024504639 hasConceptScore W2024504639C196216189 @default.
- W2024504639 hasConceptScore W2024504639C2776061190 @default.
- W2024504639 hasConceptScore W2024504639C28490314 @default.
- W2024504639 hasConceptScore W2024504639C33923547 @default.
- W2024504639 hasConceptScore W2024504639C41008148 @default.
- W2024504639 hasConceptScore W2024504639C47432892 @default.
- W2024504639 hasConceptScore W2024504639C49781872 @default.
- W2024504639 hasConceptScore W2024504639C61224824 @default.
- W2024504639 hasConceptScore W2024504639C76155785 @default.
- W2024504639 hasIssue "02" @default.
- W2024504639 hasLocation W20245046391 @default.
- W2024504639 hasOpenAccess W2024504639 @default.
- W2024504639 hasPrimaryLocation W20245046391 @default.
- W2024504639 hasRelatedWork W1574715033 @default.
- W2024504639 hasRelatedWork W1971035997 @default.
- W2024504639 hasRelatedWork W2067030110 @default.
- W2024504639 hasRelatedWork W2112217806 @default.
- W2024504639 hasRelatedWork W2128646167 @default.
- W2024504639 hasRelatedWork W2134958708 @default.
- W2024504639 hasRelatedWork W2168691954 @default.
- W2024504639 hasRelatedWork W2365480410 @default.
- W2024504639 hasRelatedWork W2381680010 @default.
- W2024504639 hasRelatedWork W3118716348 @default.
- W2024504639 hasVolume "2007" @default.
- W2024504639 isParatext "false" @default.
- W2024504639 isRetracted "false" @default.
- W2024504639 magId "2024504639" @default.
- W2024504639 workType "article" @default.