Matches in SemOpenAlex for { <https://semopenalex.org/work/W3023651494> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W3023651494 abstract "Deep neural network with dual-path bi-directional long short-term memory (BiLSTM) block has been proved to be very effective in sequence modeling, especially in speech separation, e.g. DPRNN-TasNet cite{luo2019dual}. In this paper, we propose several improvements of dual-path BiLSTM based network for end-to-end approach to monaural speech separation, which consists of 1) dual-path network with intra-parallel BiLSTM and inter-parallel BiLSTM components, 2) global context aware inter-intra cross-parallel BiLSTM, 3) multiple spiral iterative refinement dual-path BiLSTM, that all these networks take the mixed utterance of two speakers and map it to two separated utterances, where each utterance contains only one speaker's voice. For the objective, we propose to train the network by directly optimizing utterance level scale-invariant signal-to-distortion ratio (SI-SDR) in a permutation invariant training (PIT) style. Our experiments on the public WSJ0-2mix data corpus results in 20.55dB SDR improvement, 20.35dB SI-SDR improvement, 3.69 of PESQ, and 94.86% of ESTOI, which shows our proposed networks can lead to performance improvement on the speaker separation task. We have open-sourced our re-implementation of the DPRNN-TasNet in this https URL, and our LaFurca is realized based on this implementation of DPRNN-TasNet, it is believed that the results in this paper can be smoothly reproduced." @default.
- W3023651494 created "2020-05-13" @default.
- W3023651494 creator A5016279934 @default.
- W3023651494 creator A5018232553 @default.
- W3023651494 creator A5044324493 @default.
- W3023651494 date "2020-01-23" @default.
- W3023651494 modified "2023-09-23" @default.
- W3023651494 title "LaFurca: Iterative Multi-Stage Refined End-to-End Monaural Speech Separation Based on Context-Aware Dual-Path Deep Parallel Inter-Intra Bi-LSTM" @default.
- W3023651494 hasPublicationYear "2020" @default.
- W3023651494 type Work @default.
- W3023651494 sameAs 3023651494 @default.
- W3023651494 citedByCount "0" @default.
- W3023651494 crossrefType "posted-content" @default.
- W3023651494 hasAuthorship W3023651494A5016279934 @default.
- W3023651494 hasAuthorship W3023651494A5018232553 @default.
- W3023651494 hasAuthorship W3023651494A5044324493 @default.
- W3023651494 hasConcept C102894143 @default.
- W3023651494 hasConcept C103734657 @default.
- W3023651494 hasConcept C124952713 @default.
- W3023651494 hasConcept C142362112 @default.
- W3023651494 hasConcept C147168706 @default.
- W3023651494 hasConcept C151730666 @default.
- W3023651494 hasConcept C153180895 @default.
- W3023651494 hasConcept C154945302 @default.
- W3023651494 hasConcept C163294075 @default.
- W3023651494 hasConcept C2775852435 @default.
- W3023651494 hasConcept C2776182073 @default.
- W3023651494 hasConcept C2777735758 @default.
- W3023651494 hasConcept C2779343474 @default.
- W3023651494 hasConcept C2780980858 @default.
- W3023651494 hasConcept C28490314 @default.
- W3023651494 hasConcept C31258907 @default.
- W3023651494 hasConcept C41008148 @default.
- W3023651494 hasConcept C50644808 @default.
- W3023651494 hasConcept C74296488 @default.
- W3023651494 hasConcept C86803240 @default.
- W3023651494 hasConceptScore W3023651494C102894143 @default.
- W3023651494 hasConceptScore W3023651494C103734657 @default.
- W3023651494 hasConceptScore W3023651494C124952713 @default.
- W3023651494 hasConceptScore W3023651494C142362112 @default.
- W3023651494 hasConceptScore W3023651494C147168706 @default.
- W3023651494 hasConceptScore W3023651494C151730666 @default.
- W3023651494 hasConceptScore W3023651494C153180895 @default.
- W3023651494 hasConceptScore W3023651494C154945302 @default.
- W3023651494 hasConceptScore W3023651494C163294075 @default.
- W3023651494 hasConceptScore W3023651494C2775852435 @default.
- W3023651494 hasConceptScore W3023651494C2776182073 @default.
- W3023651494 hasConceptScore W3023651494C2777735758 @default.
- W3023651494 hasConceptScore W3023651494C2779343474 @default.
- W3023651494 hasConceptScore W3023651494C2780980858 @default.
- W3023651494 hasConceptScore W3023651494C28490314 @default.
- W3023651494 hasConceptScore W3023651494C31258907 @default.
- W3023651494 hasConceptScore W3023651494C41008148 @default.
- W3023651494 hasConceptScore W3023651494C50644808 @default.
- W3023651494 hasConceptScore W3023651494C74296488 @default.
- W3023651494 hasConceptScore W3023651494C86803240 @default.
- W3023651494 hasLocation W30236514941 @default.
- W3023651494 hasOpenAccess W3023651494 @default.
- W3023651494 hasPrimaryLocation W30236514941 @default.
- W3023651494 hasRelatedWork W2799930831 @default.
- W3023651494 hasRelatedWork W2800664709 @default.
- W3023651494 hasRelatedWork W2911371112 @default.
- W3023651494 hasRelatedWork W2913178639 @default.
- W3023651494 hasRelatedWork W2913914508 @default.
- W3023651494 hasRelatedWork W2935846637 @default.
- W3023651494 hasRelatedWork W2939518062 @default.
- W3023651494 hasRelatedWork W2962715207 @default.
- W3023651494 hasRelatedWork W2994988791 @default.
- W3023651494 hasRelatedWork W3001498117 @default.
- W3023651494 hasRelatedWork W3016048045 @default.
- W3023651494 hasRelatedWork W3027465796 @default.
- W3023651494 hasRelatedWork W3045904949 @default.
- W3023651494 hasRelatedWork W3084117626 @default.
- W3023651494 hasRelatedWork W3090699804 @default.
- W3023651494 hasRelatedWork W3095442114 @default.
- W3023651494 hasRelatedWork W3096796823 @default.
- W3023651494 hasRelatedWork W3131813543 @default.
- W3023651494 hasRelatedWork W3133834828 @default.
- W3023651494 hasRelatedWork W3161201807 @default.
- W3023651494 isParatext "false" @default.
- W3023651494 isRetracted "false" @default.
- W3023651494 magId "3023651494" @default.
- W3023651494 workType "article" @default.