Matches in SemOpenAlex for { <https://semopenalex.org/work/W2898721048> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W2898721048 abstract "Gaussian process (GP) audio source separation is a time-domain approach that circumvents the inherent phase approximation issue of spectrogram based methods. Furthermore, through its kernel, GPs elegantly incorporate prior knowledge about the sources into the separation model. Despite these compelling advantages, the computational complexity of GP inference scales cubically with the number of audio samples. As a result, source separation GP models have been restricted to the analysis of short audio frames. We introduce an efficient application of GPs to time-domain audio source separation, without compromising performance. For this purpose, we used GP regression, together with spectral mixture kernels, and variational sparse GPs. We compared our method with LD-PSDTF (positive semi-definite tensor factorization), KL-NMF (Kullback-Leibler non-negative matrix factorization), and IS-NMF (Itakura-Saito NMF). Results show that the proposed method outperforms these techniques." @default.
- W2898721048 created "2018-11-09" @default.
- W2898721048 creator A5005866826 @default.
- W2898721048 creator A5007248971 @default.
- W2898721048 creator A5091310279 @default.
- W2898721048 date "2018-10-30" @default.
- W2898721048 modified "2023-09-23" @default.
- W2898721048 title "Sparse Gaussian Process Audio Source Separation Using Spectrum Priors in the Time-Domain" @default.
- W2898721048 cites W137285897 @default.
- W2898721048 cites W2096354165 @default.
- W2898721048 cites W2115144768 @default.
- W2898721048 cites W2146988299 @default.
- W2898721048 cites W2149273154 @default.
- W2898721048 cites W2551943286 @default.
- W2898721048 cites W2617524636 @default.
- W2898721048 cites W2951686320 @default.
- W2898721048 cites W2952677397 @default.
- W2898721048 cites W3143596294 @default.
- W2898721048 cites W574903211 @default.
- W2898721048 cites W14377099 @default.
- W2898721048 doi "https://doi.org/10.48550/arxiv.1810.12679" @default.
- W2898721048 hasPublicationYear "2018" @default.
- W2898721048 type Work @default.
- W2898721048 sameAs 2898721048 @default.
- W2898721048 citedByCount "0" @default.
- W2898721048 crossrefType "posted-content" @default.
- W2898721048 hasAuthorship W2898721048A5005866826 @default.
- W2898721048 hasAuthorship W2898721048A5007248971 @default.
- W2898721048 hasAuthorship W2898721048A5091310279 @default.
- W2898721048 hasBestOaLocation W28987210481 @default.
- W2898721048 hasConcept C11413529 @default.
- W2898721048 hasConcept C114614502 @default.
- W2898721048 hasConcept C121332964 @default.
- W2898721048 hasConcept C152671427 @default.
- W2898721048 hasConcept C153180895 @default.
- W2898721048 hasConcept C154945302 @default.
- W2898721048 hasConcept C155281189 @default.
- W2898721048 hasConcept C158693339 @default.
- W2898721048 hasConcept C163716315 @default.
- W2898721048 hasConcept C202444582 @default.
- W2898721048 hasConcept C2776864781 @default.
- W2898721048 hasConcept C28490314 @default.
- W2898721048 hasConcept C33923547 @default.
- W2898721048 hasConcept C41008148 @default.
- W2898721048 hasConcept C42355184 @default.
- W2898721048 hasConcept C45273575 @default.
- W2898721048 hasConcept C61326573 @default.
- W2898721048 hasConcept C62520636 @default.
- W2898721048 hasConcept C74193536 @default.
- W2898721048 hasConceptScore W2898721048C11413529 @default.
- W2898721048 hasConceptScore W2898721048C114614502 @default.
- W2898721048 hasConceptScore W2898721048C121332964 @default.
- W2898721048 hasConceptScore W2898721048C152671427 @default.
- W2898721048 hasConceptScore W2898721048C153180895 @default.
- W2898721048 hasConceptScore W2898721048C154945302 @default.
- W2898721048 hasConceptScore W2898721048C155281189 @default.
- W2898721048 hasConceptScore W2898721048C158693339 @default.
- W2898721048 hasConceptScore W2898721048C163716315 @default.
- W2898721048 hasConceptScore W2898721048C202444582 @default.
- W2898721048 hasConceptScore W2898721048C2776864781 @default.
- W2898721048 hasConceptScore W2898721048C28490314 @default.
- W2898721048 hasConceptScore W2898721048C33923547 @default.
- W2898721048 hasConceptScore W2898721048C41008148 @default.
- W2898721048 hasConceptScore W2898721048C42355184 @default.
- W2898721048 hasConceptScore W2898721048C45273575 @default.
- W2898721048 hasConceptScore W2898721048C61326573 @default.
- W2898721048 hasConceptScore W2898721048C62520636 @default.
- W2898721048 hasConceptScore W2898721048C74193536 @default.
- W2898721048 hasLocation W28987210481 @default.
- W2898721048 hasLocation W28987210482 @default.
- W2898721048 hasLocation W28987210483 @default.
- W2898721048 hasOpenAccess W2898721048 @default.
- W2898721048 hasPrimaryLocation W28987210481 @default.
- W2898721048 hasRelatedWork W1984517456 @default.
- W2898721048 hasRelatedWork W2058269967 @default.
- W2898721048 hasRelatedWork W2081322759 @default.
- W2898721048 hasRelatedWork W2098101267 @default.
- W2898721048 hasRelatedWork W2321942335 @default.
- W2898721048 hasRelatedWork W2339777604 @default.
- W2898721048 hasRelatedWork W2576645943 @default.
- W2898721048 hasRelatedWork W2615370628 @default.
- W2898721048 hasRelatedWork W2804271779 @default.
- W2898721048 hasRelatedWork W2898721048 @default.
- W2898721048 isParatext "false" @default.
- W2898721048 isRetracted "false" @default.
- W2898721048 magId "2898721048" @default.
- W2898721048 workType "article" @default.