Matches in SemOpenAlex for { <https://semopenalex.org/work/W3117310166> ?p ?o ?g. }
- W3117310166 endingPage "228753" @default.
- W3117310166 startingPage "228740" @default.
- W3117310166 abstract "This paper proposes a fast optimization algorithm for the multichannel variational autoencoder (MVAE) method, a recently proposed powerful multichannel source separation technique. The MVAE method can achieve good source separation performance thanks to a convergence-guaranteed optimization algorithm and the idea of jointly performing multi-speaker separation and speaker identification. However, one drawback is the high computational cost of the optimization algorithm. To overcome this drawback, this paper proposes using an auxiliary classifier VAE, an information-theoretic extension of the conditional VAE (CVAE), to train the generative model of the source spectrograms and using it to efficiently update the parameters of the source spectrogram models at each iteration of the source separation algorithm. We call the proposed algorithm “FastMVAE” (or fMVAE for short). Experimental evaluations revealed that the proposed fast algorithm can achieve high source separation performance in both speaker-dependent and speaker-independent scenarios while significantly reducing the computational time compared to the original MVAE method by more than 90% on both GPU and CPU. However, there is still room for improvement of about 3 dB compared to the original MVAE method." @default.
- W3117310166 created "2021-01-05" @default.
- W3117310166 creator A5001243214 @default.
- W3117310166 creator A5051093194 @default.
- W3117310166 creator A5061981123 @default.
- W3117310166 creator A5075702573 @default.
- W3117310166 date "2020-01-01" @default.
- W3117310166 modified "2023-09-24" @default.
- W3117310166 title "FastMVAE: A Fast Optimization Algorithm for the Multichannel Variational Autoencoder Method" @default.
- W3117310166 cites W1552314771 @default.
- W3117310166 cites W2058341666 @default.
- W3117310166 cites W2067295501 @default.
- W3117310166 cites W2072548008 @default.
- W3117310166 cites W2086286498 @default.
- W3117310166 cites W2113990625 @default.
- W3117310166 cites W2116064496 @default.
- W3117310166 cites W2117678320 @default.
- W3117310166 cites W2127851351 @default.
- W3117310166 cites W2130416410 @default.
- W3117310166 cites W2139302694 @default.
- W3117310166 cites W2149414429 @default.
- W3117310166 cites W2168273590 @default.
- W3117310166 cites W2221409856 @default.
- W3117310166 cites W2408744528 @default.
- W3117310166 cites W2412956798 @default.
- W3117310166 cites W2460742184 @default.
- W3117310166 cites W2734774145 @default.
- W3117310166 cites W2747744257 @default.
- W3117310166 cites W2750380122 @default.
- W3117310166 cites W2750446090 @default.
- W3117310166 cites W2766672686 @default.
- W3117310166 cites W2792498316 @default.
- W3117310166 cites W2804644188 @default.
- W3117310166 cites W2885661542 @default.
- W3117310166 cites W2892163332 @default.
- W3117310166 cites W2894785362 @default.
- W3117310166 cites W2901552243 @default.
- W3117310166 cites W2905196628 @default.
- W3117310166 cites W2911579794 @default.
- W3117310166 cites W2922004249 @default.
- W3117310166 cites W2936446744 @default.
- W3117310166 cites W2946555236 @default.
- W3117310166 cites W2954049404 @default.
- W3117310166 cites W2962715207 @default.
- W3117310166 cites W2962866211 @default.
- W3117310166 cites W2963341071 @default.
- W3117310166 cites W2963375116 @default.
- W3117310166 cites W2963969588 @default.
- W3117310166 cites W2968723745 @default.
- W3117310166 cites W2972460025 @default.
- W3117310166 cites W2972516210 @default.
- W3117310166 cites W2979850772 @default.
- W3117310166 cites W2989942677 @default.
- W3117310166 doi "https://doi.org/10.1109/access.2020.3045704" @default.
- W3117310166 hasPublicationYear "2020" @default.
- W3117310166 type Work @default.
- W3117310166 sameAs 3117310166 @default.
- W3117310166 citedByCount "4" @default.
- W3117310166 countsByYear W31173101662021 @default.
- W3117310166 countsByYear W31173101662022 @default.
- W3117310166 countsByYear W31173101662023 @default.
- W3117310166 crossrefType "journal-article" @default.
- W3117310166 hasAuthorship W3117310166A5001243214 @default.
- W3117310166 hasAuthorship W3117310166A5051093194 @default.
- W3117310166 hasAuthorship W3117310166A5061981123 @default.
- W3117310166 hasAuthorship W3117310166A5075702573 @default.
- W3117310166 hasBestOaLocation W31173101661 @default.
- W3117310166 hasConcept C101738243 @default.
- W3117310166 hasConcept C11413529 @default.
- W3117310166 hasConcept C126255220 @default.
- W3117310166 hasConcept C154945302 @default.
- W3117310166 hasConcept C2987595161 @default.
- W3117310166 hasConcept C33923547 @default.
- W3117310166 hasConcept C41008148 @default.
- W3117310166 hasConcept C50644808 @default.
- W3117310166 hasConceptScore W3117310166C101738243 @default.
- W3117310166 hasConceptScore W3117310166C11413529 @default.
- W3117310166 hasConceptScore W3117310166C126255220 @default.
- W3117310166 hasConceptScore W3117310166C154945302 @default.
- W3117310166 hasConceptScore W3117310166C2987595161 @default.
- W3117310166 hasConceptScore W3117310166C33923547 @default.
- W3117310166 hasConceptScore W3117310166C41008148 @default.
- W3117310166 hasConceptScore W3117310166C50644808 @default.
- W3117310166 hasFunder F4320334764 @default.
- W3117310166 hasFunder F4320338075 @default.
- W3117310166 hasLocation W31173101661 @default.
- W3117310166 hasLocation W31173101662 @default.
- W3117310166 hasOpenAccess W3117310166 @default.
- W3117310166 hasPrimaryLocation W31173101661 @default.
- W3117310166 hasRelatedWork W2283340597 @default.
- W3117310166 hasRelatedWork W2351491280 @default.
- W3117310166 hasRelatedWork W2371447506 @default.
- W3117310166 hasRelatedWork W2386767533 @default.
- W3117310166 hasRelatedWork W2785529134 @default.
- W3117310166 hasRelatedWork W2927931735 @default.
- W3117310166 hasRelatedWork W2946739205 @default.
- W3117310166 hasRelatedWork W3019797369 @default.
- W3117310166 hasRelatedWork W303980170 @default.