Matches in SemOpenAlex for { <https://semopenalex.org/work/W2998302919> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2998302919 abstract "Utterance-level permutation invariant training (uPIT) has achieved promising progress on single-channel multi-talker speech separation task. Long short-term memory (LSTM) and bidirectional LSTM (BLSTM) are widely used as the separation networks of uPIT, i.e. uPIT-LSTM and uPIT-BLSTM. uPIT-LSTM has lower latency but worse performance, while uPIT-BLSTM has better performance but higher latency. In this paper, we propose using latency-controlled BLSTM (LC-BLSTM) during inference to fulfill low-latency and good-performance speech separation. To find a better training strategy for BLSTM-based separation network, chunk-level PIT (cPIT) and uPIT are compared. The experimental results show that uPIT outperforms cPIT when LC-BLSTM is used during inference. It is also found that the inter-chunk speaker tracing (ST) can further improve the separation performance of uPIT-LC-BLSTM. Evaluated on the WSJ0 two-talker mixed-speech separation task, the absolute gap of signal-to-distortion ratio (SDR) between uPIT-BLSTM and uPIT-LC-BLSTM is reduced to within 0.7 dB." @default.
- W2998302919 created "2020-01-10" @default.
- W2998302919 creator A5033070203 @default.
- W2998302919 creator A5036446253 @default.
- W2998302919 creator A5049134051 @default.
- W2998302919 creator A5065666608 @default.
- W2998302919 creator A5068911982 @default.
- W2998302919 creator A5072887461 @default.
- W2998302919 date "2019-12-25" @default.
- W2998302919 modified "2023-09-26" @default.
- W2998302919 title "Utterance-level Permutation Invariant Training with Latency-controlled BLSTM for Single-channel Multi-talker Speech Separation" @default.
- W2998302919 cites W1665214252 @default.
- W2998302919 cites W185399533 @default.
- W2998302919 cites W2064675550 @default.
- W2998302919 cites W2143169494 @default.
- W2998302919 cites W2143612262 @default.
- W2998302919 cites W2147568880 @default.
- W2998302919 cites W2221409856 @default.
- W2998302919 cites W2329068866 @default.
- W2998302919 cites W2460742184 @default.
- W2998302919 cites W2533523411 @default.
- W2998302919 cites W2561557072 @default.
- W2998302919 cites W2631415506 @default.
- W2998302919 cites W2729190387 @default.
- W2998302919 cites W2800022361 @default.
- W2998302919 cites W2888957211 @default.
- W2998302919 cites W2889010580 @default.
- W2998302919 cites W2889540509 @default.
- W2998302919 cites W2892025684 @default.
- W2998302919 cites W2916979304 @default.
- W2998302919 cites W2962935966 @default.
- W2998302919 cites W2963574857 @default.
- W2998302919 cites W2963773971 @default.
- W2998302919 cites W2963843276 @default.
- W2998302919 cites W2964084166 @default.
- W2998302919 cites W3099832538 @default.
- W2998302919 cites W2890505457 @default.
- W2998302919 cites W3023071679 @default.
- W2998302919 doi "https://doi.org/10.48550/arxiv.1912.11613" @default.
- W2998302919 hasPublicationYear "2019" @default.
- W2998302919 type Work @default.
- W2998302919 sameAs 2998302919 @default.
- W2998302919 citedByCount "0" @default.
- W2998302919 crossrefType "posted-content" @default.
- W2998302919 hasAuthorship W2998302919A5033070203 @default.
- W2998302919 hasAuthorship W2998302919A5036446253 @default.
- W2998302919 hasAuthorship W2998302919A5049134051 @default.
- W2998302919 hasAuthorship W2998302919A5065666608 @default.
- W2998302919 hasAuthorship W2998302919A5068911982 @default.
- W2998302919 hasAuthorship W2998302919A5072887461 @default.
- W2998302919 hasBestOaLocation W29983029191 @default.
- W2998302919 hasConcept C154945302 @default.
- W2998302919 hasConcept C2775852435 @default.
- W2998302919 hasConcept C2776214188 @default.
- W2998302919 hasConcept C28490314 @default.
- W2998302919 hasConcept C41008148 @default.
- W2998302919 hasConcept C76155785 @default.
- W2998302919 hasConcept C82876162 @default.
- W2998302919 hasConceptScore W2998302919C154945302 @default.
- W2998302919 hasConceptScore W2998302919C2775852435 @default.
- W2998302919 hasConceptScore W2998302919C2776214188 @default.
- W2998302919 hasConceptScore W2998302919C28490314 @default.
- W2998302919 hasConceptScore W2998302919C41008148 @default.
- W2998302919 hasConceptScore W2998302919C76155785 @default.
- W2998302919 hasConceptScore W2998302919C82876162 @default.
- W2998302919 hasLocation W29983029191 @default.
- W2998302919 hasLocation W29983029192 @default.
- W2998302919 hasOpenAccess W2998302919 @default.
- W2998302919 hasPrimaryLocation W29983029191 @default.
- W2998302919 hasRelatedWork W1596769518 @default.
- W2998302919 hasRelatedWork W2147934395 @default.
- W2998302919 hasRelatedWork W2886693075 @default.
- W2998302919 hasRelatedWork W3184187848 @default.
- W2998302919 hasRelatedWork W3197304116 @default.
- W2998302919 hasRelatedWork W3198503472 @default.
- W2998302919 hasRelatedWork W4226175622 @default.
- W2998302919 hasRelatedWork W4287645148 @default.
- W2998302919 hasRelatedWork W4327525618 @default.
- W2998302919 hasRelatedWork W2779562428 @default.
- W2998302919 isParatext "false" @default.
- W2998302919 isRetracted "false" @default.
- W2998302919 magId "2998302919" @default.
- W2998302919 workType "article" @default.