Matches in SemOpenAlex for { <https://semopenalex.org/work/W2801356078> ?p ?o ?g. }
- W2801356078 abstract "Convolutional neural networks have recently been found useful in music information retrieval tasks such as genre classification, music recommendation, and instrument detection. In these deep neural network applications, music data is often converted into a spectrogram and treated as a single-channel image input. In this work we enrich the semantic musical understanding of the network by using source separation to separate the single channel into multiple, meaningful channels. As a result, we improve accuracy and accelerate training. We evaluate this method in a genre classification task with a variety of CNN architectures. Compared to single-channel inputs, we achieve an average of a 2.7% increase in accuracy across all trained models using a 3-channel source-separated model and a 2.2% increase using a 4-channel model. This is due to quicker convergence of the loss function." @default.
- W2801356078 created "2018-05-17" @default.
- W2801356078 creator A5032475867 @default.
- W2801356078 creator A5064808783 @default.
- W2801356078 creator A5089506619 @default.
- W2801356078 date "2017-11-01" @default.
- W2801356078 modified "2023-09-24" @default.
- W2801356078 title "Source-separated audio input for accelerating convolutional neural networks" @default.
- W2801356078 cites W1524450315 @default.
- W2801356078 cites W1556219185 @default.
- W2801356078 cites W1686810756 @default.
- W2801356078 cites W1981755271 @default.
- W2801356078 cites W2042390666 @default.
- W2801356078 cites W2053165762 @default.
- W2801356078 cites W2103869314 @default.
- W2801356078 cites W2107789863 @default.
- W2801356078 cites W2127870748 @default.
- W2801356078 cites W2133824856 @default.
- W2801356078 cites W2137028279 @default.
- W2801356078 cites W2155273149 @default.
- W2801356078 cites W2163605009 @default.
- W2801356078 cites W2178339699 @default.
- W2801356078 cites W2196753680 @default.
- W2801356078 cites W2257979135 @default.
- W2801356078 cites W2399621629 @default.
- W2801356078 cites W2404620398 @default.
- W2801356078 cites W2404724332 @default.
- W2801356078 cites W2437181147 @default.
- W2801356078 cites W2563215737 @default.
- W2801356078 cites W2949117887 @default.
- W2801356078 cites W3098670224 @default.
- W2801356078 doi "https://doi.org/10.1109/wnyipw.2017.8356323" @default.
- W2801356078 hasPublicationYear "2017" @default.
- W2801356078 type Work @default.
- W2801356078 sameAs 2801356078 @default.
- W2801356078 citedByCount "0" @default.
- W2801356078 crossrefType "proceedings-article" @default.
- W2801356078 hasAuthorship W2801356078A5032475867 @default.
- W2801356078 hasAuthorship W2801356078A5064808783 @default.
- W2801356078 hasAuthorship W2801356078A5089506619 @default.
- W2801356078 hasConcept C108583219 @default.
- W2801356078 hasConcept C119857082 @default.
- W2801356078 hasConcept C127162648 @default.
- W2801356078 hasConcept C153180895 @default.
- W2801356078 hasConcept C154945302 @default.
- W2801356078 hasConcept C162324750 @default.
- W2801356078 hasConcept C187736073 @default.
- W2801356078 hasConcept C2776864781 @default.
- W2801356078 hasConcept C2777303404 @default.
- W2801356078 hasConcept C2780451532 @default.
- W2801356078 hasConcept C28490314 @default.
- W2801356078 hasConcept C41008148 @default.
- W2801356078 hasConcept C45273575 @default.
- W2801356078 hasConcept C50522688 @default.
- W2801356078 hasConcept C50644808 @default.
- W2801356078 hasConcept C76155785 @default.
- W2801356078 hasConcept C81363708 @default.
- W2801356078 hasConceptScore W2801356078C108583219 @default.
- W2801356078 hasConceptScore W2801356078C119857082 @default.
- W2801356078 hasConceptScore W2801356078C127162648 @default.
- W2801356078 hasConceptScore W2801356078C153180895 @default.
- W2801356078 hasConceptScore W2801356078C154945302 @default.
- W2801356078 hasConceptScore W2801356078C162324750 @default.
- W2801356078 hasConceptScore W2801356078C187736073 @default.
- W2801356078 hasConceptScore W2801356078C2776864781 @default.
- W2801356078 hasConceptScore W2801356078C2777303404 @default.
- W2801356078 hasConceptScore W2801356078C2780451532 @default.
- W2801356078 hasConceptScore W2801356078C28490314 @default.
- W2801356078 hasConceptScore W2801356078C41008148 @default.
- W2801356078 hasConceptScore W2801356078C45273575 @default.
- W2801356078 hasConceptScore W2801356078C50522688 @default.
- W2801356078 hasConceptScore W2801356078C50644808 @default.
- W2801356078 hasConceptScore W2801356078C76155785 @default.
- W2801356078 hasConceptScore W2801356078C81363708 @default.
- W2801356078 hasLocation W28013560781 @default.
- W2801356078 hasOpenAccess W2801356078 @default.
- W2801356078 hasPrimaryLocation W28013560781 @default.
- W2801356078 hasRelatedWork W2593179621 @default.
- W2801356078 hasRelatedWork W2775123266 @default.
- W2801356078 hasRelatedWork W2777468850 @default.
- W2801356078 hasRelatedWork W2794150026 @default.
- W2801356078 hasRelatedWork W2804960024 @default.
- W2801356078 hasRelatedWork W2807116244 @default.
- W2801356078 hasRelatedWork W2897725921 @default.
- W2801356078 hasRelatedWork W2903594313 @default.
- W2801356078 hasRelatedWork W2935162632 @default.
- W2801356078 hasRelatedWork W2935707851 @default.
- W2801356078 hasRelatedWork W2945399614 @default.
- W2801356078 hasRelatedWork W2963075964 @default.
- W2801356078 hasRelatedWork W2963517948 @default.
- W2801356078 hasRelatedWork W2963751183 @default.
- W2801356078 hasRelatedWork W2972620405 @default.
- W2801356078 hasRelatedWork W3004635547 @default.
- W2801356078 hasRelatedWork W3034207266 @default.
- W2801356078 hasRelatedWork W3122728801 @default.
- W2801356078 hasRelatedWork W3133795225 @default.
- W2801356078 hasRelatedWork W3191401413 @default.
- W2801356078 isParatext "false" @default.
- W2801356078 isRetracted "false" @default.
- W2801356078 magId "2801356078" @default.