Matches in SemOpenAlex for { <https://semopenalex.org/work/W4242220767> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4242220767 abstract "Current augmented Kalman filter (AKF)-based speech enhancement algorithms utilise a temporal convolutional network (TCN) to estimate the clean speech and noise linear prediction coefficient (LPC). However, the multi-head attention network (MHANet) has demonstrated the ability to more efficiently model the long-term dependencies of noisy speech than TCNs. Motivated by this, we investigate the MHANet for LPC estimation. We aim to produce clean speech and noise LPC parameters with the least bias to date. With this, we also aim to produce higher quality and more intelligible enhanced speech than any current KF or AKF-based SEA. Here, we investigate MHANet within the DeepLPC framework. DeepLPC is a deep learning framework for jointly estimating the clean speech and noise LPC power spectra. DeepLPC is selected as it exhibits significantly less bias than other frameworks, by avoiding the use of whitening filters and post-processing. DeepLPC-MHANet is evaluated on the NOIZEUS corpus using subjective AB listening tests, as well as seven different objective measures (CSIG, CBAK, COVL, PESQ, STOI, SegSNR, and SI-SDR). DeepLPC-MHANet is compared to five existing deep learning-based methods. Compared to other deep learning approaches, DeepLPC-MHANet produced clean speech LPC estimates with the least amount of bias. DeepLPC-MHANet-AKF also produced higher objective scores than any of the competing methods (with an improvement of 0.17 for CSIG, 0.15 for CBAK, 0.19 for COVL, 0.24 for PESQ, 3.70% for STOI, 1.03 dB for SegSNR, and 1.04 dB for SI-SDR over the next best method). The enhanced speech produced by DeepLPC-MHANet-AKF was also the most preferred amongst ten listeners. By producing LPC estimates with the least amount of bias to date, DeepLPC-MHANet enables the AKF to produce enhanced speech at a higher quality and intelligibility than any previous method." @default.
- W4242220767 created "2022-05-12" @default.
- W4242220767 creator A5000160818 @default.
- W4242220767 creator A5041494929 @default.
- W4242220767 creator A5046089538 @default.
- W4242220767 date "2021-04-12" @default.
- W4242220767 modified "2023-09-29" @default.
- W4242220767 title "DeepLPC-MHANet: Multi-Head Self-Attention for Augmented Kalman Filter-based Speech Enhancement" @default.
- W4242220767 doi "https://doi.org/10.36227/techrxiv.14384909.v1" @default.
- W4242220767 hasPublicationYear "2021" @default.
- W4242220767 type Work @default.
- W4242220767 citedByCount "0" @default.
- W4242220767 crossrefType "posted-content" @default.
- W4242220767 hasAuthorship W4242220767A5000160818 @default.
- W4242220767 hasAuthorship W4242220767A5041494929 @default.
- W4242220767 hasAuthorship W4242220767A5046089538 @default.
- W4242220767 hasBestOaLocation W42422207671 @default.
- W4242220767 hasConcept C103734657 @default.
- W4242220767 hasConcept C115961682 @default.
- W4242220767 hasConcept C13280743 @default.
- W4242220767 hasConcept C154945302 @default.
- W4242220767 hasConcept C157286648 @default.
- W4242220767 hasConcept C15744967 @default.
- W4242220767 hasConcept C163294075 @default.
- W4242220767 hasConcept C177291462 @default.
- W4242220767 hasConcept C185798385 @default.
- W4242220767 hasConcept C205649164 @default.
- W4242220767 hasConcept C2776182073 @default.
- W4242220767 hasConcept C28490314 @default.
- W4242220767 hasConcept C41008148 @default.
- W4242220767 hasConcept C46312422 @default.
- W4242220767 hasConcept C99498987 @default.
- W4242220767 hasConceptScore W4242220767C103734657 @default.
- W4242220767 hasConceptScore W4242220767C115961682 @default.
- W4242220767 hasConceptScore W4242220767C13280743 @default.
- W4242220767 hasConceptScore W4242220767C154945302 @default.
- W4242220767 hasConceptScore W4242220767C157286648 @default.
- W4242220767 hasConceptScore W4242220767C15744967 @default.
- W4242220767 hasConceptScore W4242220767C163294075 @default.
- W4242220767 hasConceptScore W4242220767C177291462 @default.
- W4242220767 hasConceptScore W4242220767C185798385 @default.
- W4242220767 hasConceptScore W4242220767C205649164 @default.
- W4242220767 hasConceptScore W4242220767C2776182073 @default.
- W4242220767 hasConceptScore W4242220767C28490314 @default.
- W4242220767 hasConceptScore W4242220767C41008148 @default.
- W4242220767 hasConceptScore W4242220767C46312422 @default.
- W4242220767 hasConceptScore W4242220767C99498987 @default.
- W4242220767 hasLocation W42422207671 @default.
- W4242220767 hasLocation W42422207672 @default.
- W4242220767 hasLocation W42422207673 @default.
- W4242220767 hasOpenAccess W4242220767 @default.
- W4242220767 hasPrimaryLocation W42422207671 @default.
- W4242220767 hasRelatedWork W10351748 @default.
- W4242220767 hasRelatedWork W12405394 @default.
- W4242220767 hasRelatedWork W14578209 @default.
- W4242220767 hasRelatedWork W3703470 @default.
- W4242220767 hasRelatedWork W589191 @default.
- W4242220767 hasRelatedWork W6558424 @default.
- W4242220767 hasRelatedWork W7128961 @default.
- W4242220767 hasRelatedWork W8932171 @default.
- W4242220767 hasRelatedWork W9420449 @default.
- W4242220767 hasRelatedWork W9997782 @default.
- W4242220767 isParatext "false" @default.
- W4242220767 isRetracted "false" @default.
- W4242220767 workType "article" @default.