Matches in SemOpenAlex for { <https://semopenalex.org/work/W2972346374> ?p ?o ?g. }
- W2972346374 abstract "In this paper, we propose a novel end-to-end neural-network-based speaker diarization method. Unlike most existing methods, our proposed method does not have separate modules for extraction and clustering of speaker representations. Instead, our model has a single neural network that directly outputs speaker diarization results. To realize such a model, we formulate the speaker diarization problem as a multi-label classification problem, and introduces a permutation-free objective function to directly minimize diarization errors without being suffered from the speaker-label permutation problem. Besides its end-to-end simplicity, the proposed method also benefits from being able to explicitly handle overlapping speech during training and inference. Because of the benefit, our model can be easily trained/adapted with real-recorded multi-speaker conversations just by feeding the corresponding multi-speaker segment labels. We evaluated the proposed method on simulated speech mixtures. The proposed method achieved diarization error rate of 12.28%, while a conventional clustering-based system produced diarization error rate of 28.77%. Furthermore, the domain adaptation with real-recorded speech provided 25.6% relative improvement on the CALLHOME dataset. Our source code is available online at https://github.com/hitachi-speech/EEND." @default.
- W2972346374 created "2019-09-19" @default.
- W2972346374 creator A5001291873 @default.
- W2972346374 creator A5016279564 @default.
- W2972346374 creator A5026324656 @default.
- W2972346374 creator A5044818016 @default.
- W2972346374 creator A5076987349 @default.
- W2972346374 date "2019-09-12" @default.
- W2972346374 modified "2023-10-09" @default.
- W2972346374 title "End-to-End Neural Speaker Diarization with Permutation-Free Objectives" @default.
- W2972346374 cites W1524333225 @default.
- W2972346374 cites W1591607137 @default.
- W2972346374 cites W2038101708 @default.
- W2972346374 cites W2081074144 @default.
- W2972346374 cites W2093499222 @default.
- W2972346374 cites W2150769028 @default.
- W2972346374 cites W2159591770 @default.
- W2972346374 cites W2170579896 @default.
- W2972346374 cites W2219249508 @default.
- W2972346374 cites W2221409856 @default.
- W2972346374 cites W2460742184 @default.
- W2972346374 cites W2620757702 @default.
- W2972346374 cites W2638067502 @default.
- W2972346374 cites W2696967604 @default.
- W2972346374 cites W2734774145 @default.
- W2972346374 cites W2746574320 @default.
- W2972346374 cites W2763761345 @default.
- W2972346374 cites W2786458517 @default.
- W2972346374 cites W2884797218 @default.
- W2972346374 cites W2889031312 @default.
- W2972346374 cites W2889381673 @default.
- W2972346374 cites W2889418727 @default.
- W2972346374 cites W2890964092 @default.
- W2972346374 cites W2891247151 @default.
- W2972346374 cites W2896538040 @default.
- W2972346374 cites W2900091092 @default.
- W2972346374 cites W2900212944 @default.
- W2972346374 cites W2900440209 @default.
- W2972346374 cites W2939690918 @default.
- W2972346374 cites W2952752702 @default.
- W2972346374 cites W2962788625 @default.
- W2972346374 cites W2963470929 @default.
- W2972346374 cites W2964121744 @default.
- W2972346374 cites W2521624909 @default.
- W2972346374 doi "https://doi.org/10.48550/arxiv.1909.05952" @default.
- W2972346374 hasPublicationYear "2019" @default.
- W2972346374 type Work @default.
- W2972346374 sameAs 2972346374 @default.
- W2972346374 citedByCount "1" @default.
- W2972346374 crossrefType "posted-content" @default.
- W2972346374 hasAuthorship W2972346374A5001291873 @default.
- W2972346374 hasAuthorship W2972346374A5016279564 @default.
- W2972346374 hasAuthorship W2972346374A5026324656 @default.
- W2972346374 hasAuthorship W2972346374A5044818016 @default.
- W2972346374 hasAuthorship W2972346374A5076987349 @default.
- W2972346374 hasBestOaLocation W29723463741 @default.
- W2972346374 hasConcept C121332964 @default.
- W2972346374 hasConcept C133892786 @default.
- W2972346374 hasConcept C149838564 @default.
- W2972346374 hasConcept C153180895 @default.
- W2972346374 hasConcept C154945302 @default.
- W2972346374 hasConcept C21308566 @default.
- W2972346374 hasConcept C24890656 @default.
- W2972346374 hasConcept C2776214188 @default.
- W2972346374 hasConcept C28490314 @default.
- W2972346374 hasConcept C40969351 @default.
- W2972346374 hasConcept C41008148 @default.
- W2972346374 hasConcept C50644808 @default.
- W2972346374 hasConcept C73555534 @default.
- W2972346374 hasConcept C74296488 @default.
- W2972346374 hasConceptScore W2972346374C121332964 @default.
- W2972346374 hasConceptScore W2972346374C133892786 @default.
- W2972346374 hasConceptScore W2972346374C149838564 @default.
- W2972346374 hasConceptScore W2972346374C153180895 @default.
- W2972346374 hasConceptScore W2972346374C154945302 @default.
- W2972346374 hasConceptScore W2972346374C21308566 @default.
- W2972346374 hasConceptScore W2972346374C24890656 @default.
- W2972346374 hasConceptScore W2972346374C2776214188 @default.
- W2972346374 hasConceptScore W2972346374C28490314 @default.
- W2972346374 hasConceptScore W2972346374C40969351 @default.
- W2972346374 hasConceptScore W2972346374C41008148 @default.
- W2972346374 hasConceptScore W2972346374C50644808 @default.
- W2972346374 hasConceptScore W2972346374C73555534 @default.
- W2972346374 hasConceptScore W2972346374C74296488 @default.
- W2972346374 hasLocation W29723463741 @default.
- W2972346374 hasOpenAccess W2972346374 @default.
- W2972346374 hasPrimaryLocation W29723463741 @default.
- W2972346374 hasRelatedWork W1521049138 @default.
- W2972346374 hasRelatedWork W1960256358 @default.
- W2972346374 hasRelatedWork W2112059504 @default.
- W2972346374 hasRelatedWork W2144208207 @default.
- W2972346374 hasRelatedWork W2499802997 @default.
- W2972346374 hasRelatedWork W2972346374 @default.
- W2972346374 hasRelatedWork W4221117560 @default.
- W2972346374 hasRelatedWork W4225792560 @default.
- W2972346374 hasRelatedWork W4225890157 @default.
- W2972346374 hasRelatedWork W4287236246 @default.
- W2972346374 isParatext "false" @default.
- W2972346374 isRetracted "false" @default.
- W2972346374 magId "2972346374" @default.