FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2902894493> ?p ?o ?g. }

• W2902894493 endingPage "43" @default.
• W2902894493 startingPage "31" @default.
• W2902894493 abstract "We propose a novel iterative mask estimation (IME) framework to improve the state-of-the-art complex Gaussian mixture model (CGMM)-based beamforming approach in an iterative manner by leveraging upon the complementary information obtained from different deep models. Although CGMM has been recently demonstrated to be quite effective for multi-channel, automation speech recognition (ASR) in operational scenarios, the corresponding mask estimation, however, is not always accurate in adverse environments due to the lack of prior or context information. To address this problem, in this study, a neural-network-based ideal ratio mask estimator learned from a multi-condition data set is first adopted to incorporate prior information, obtained from the speech/noise interactions and the long acoustic context, into CGMM-based beamformed speech that has a higher signal-to-noise ratio (SNR) than the original noisy speech signal. Next, to further utilize the rich context information in deep acoustic and language models, voice activity detection information, obtained from speech recognition results, is then used to refine mask estimation, yielding a significant reduction in insertion errors. During testing on the recently launched CHiME-4 Challenge ASR task of recognizing 6-channel microphone array speech, the proposed IME approach significantly and consistently outperforms the CGMM approach under different configurations, with relative word error rate reductions ranging from 20% to 30%. Furthermore, the IME approach plays a key role in the ensemble system that achieves the best performance in the CHiME-4 Challenge." @default.
• W2902894493 created "2018-12-11" @default.
• W2902894493 creator A5042149099 @default.
• W2902894493 creator A5049067709 @default.
• W2902894493 creator A5066595711 @default.
• W2902894493 creator A5066868860 @default.
• W2902894493 creator A5077092506 @default.
• W2902894493 creator A5077696541 @default.
• W2902894493 creator A5089891264 @default.
• W2902894493 date "2019-01-01" @default.
• W2902894493 modified "2023-10-13" @default.
• W2902894493 title "An iterative mask estimation approach to deep learning based multi-channel speech recognition" @default.
• W2902894493 cites W190004713 @default.
• W2902894493 cites W1968939597 @default.
• W2902894493 cites W1978752674 @default.
• W2902894493 cites W1993882792 @default.
• W2902894493 cites W1996355918 @default.
• W2902894493 cites W2002342963 @default.
• W2902894493 cites W2039553327 @default.
• W2902894493 cites W2048325254 @default.
• W2902894493 cites W2050875687 @default.
• W2902894493 cites W2060108923 @default.
• W2902894493 cites W2064675550 @default.
• W2902894493 cites W2066218102 @default.
• W2902894493 cites W2069884666 @default.
• W2902894493 cites W2087774313 @default.
• W2902894493 cites W2093225945 @default.
• W2902894493 cites W2099984896 @default.
• W2902894493 cites W2100969003 @default.
• W2902894493 cites W2126942983 @default.
• W2902894493 cites W2128653836 @default.
• W2902894493 cites W2129120544 @default.
• W2902894493 cites W2129934596 @default.
• W2902894493 cites W2134642725 @default.
• W2902894493 cites W2148613904 @default.
• W2902894493 cites W2160815625 @default.
• W2902894493 cites W2164570496 @default.
• W2902894493 cites W2164764235 @default.
• W2902894493 cites W2168379380 @default.
• W2902894493 cites W2345067732 @default.
• W2902894493 cites W2408744528 @default.
• W2902894493 cites W2476521382 @default.
• W2902894493 cites W2559809918 @default.
• W2902894493 cites W2589857635 @default.
• W2902894493 cites W2702006285 @default.
• W2902894493 cites W2767071179 @default.
• W2902894493 cites W3147539069 @default.
• W2902894493 doi "https://doi.org/10.1016/j.specom.2018.11.005" @default.
• W2902894493 hasPublicationYear "2019" @default.
• W2902894493 type Work @default.
• W2902894493 sameAs 2902894493 @default.
• W2902894493 citedByCount "20" @default.
• W2902894493 countsByYear W29028944932019 @default.
• W2902894493 countsByYear W29028944932020 @default.
• W2902894493 countsByYear W29028944932021 @default.
• W2902894493 countsByYear W29028944932022 @default.
• W2902894493 countsByYear W29028944932023 @default.
• W2902894493 crossrefType "journal-article" @default.
• W2902894493 hasAuthorship W2902894493A5042149099 @default.
• W2902894493 hasAuthorship W2902894493A5049067709 @default.
• W2902894493 hasAuthorship W2902894493A5066595711 @default.
• W2902894493 hasAuthorship W2902894493A5066868860 @default.
• W2902894493 hasAuthorship W2902894493A5077092506 @default.
• W2902894493 hasAuthorship W2902894493A5077696541 @default.
• W2902894493 hasAuthorship W2902894493A5089891264 @default.
• W2902894493 hasConcept C105795698 @default.
• W2902894493 hasConcept C108583219 @default.
• W2902894493 hasConcept C115961682 @default.
• W2902894493 hasConcept C127162648 @default.
• W2902894493 hasConcept C151730666 @default.
• W2902894493 hasConcept C153180895 @default.
• W2902894493 hasConcept C154945302 @default.
• W2902894493 hasConcept C163294075 @default.
• W2902894493 hasConcept C185429906 @default.
• W2902894493 hasConcept C2776182073 @default.
• W2902894493 hasConcept C2778263558 @default.
• W2902894493 hasConcept C2779343474 @default.
• W2902894493 hasConcept C28490314 @default.
• W2902894493 hasConcept C31258907 @default.
• W2902894493 hasConcept C33923547 @default.
• W2902894493 hasConcept C40969351 @default.
• W2902894493 hasConcept C41008148 @default.
• W2902894493 hasConcept C54197355 @default.
• W2902894493 hasConcept C68115822 @default.
• W2902894493 hasConcept C76155785 @default.
• W2902894493 hasConcept C86803240 @default.
• W2902894493 hasConcept C99498987 @default.
• W2902894493 hasConceptScore W2902894493C105795698 @default.
• W2902894493 hasConceptScore W2902894493C108583219 @default.
• W2902894493 hasConceptScore W2902894493C115961682 @default.
• W2902894493 hasConceptScore W2902894493C127162648 @default.
• W2902894493 hasConceptScore W2902894493C151730666 @default.
• W2902894493 hasConceptScore W2902894493C153180895 @default.
• W2902894493 hasConceptScore W2902894493C154945302 @default.
• W2902894493 hasConceptScore W2902894493C163294075 @default.
• W2902894493 hasConceptScore W2902894493C185429906 @default.
• W2902894493 hasConceptScore W2902894493C2776182073 @default.
• W2902894493 hasConceptScore W2902894493C2778263558 @default.

• next