Noise adaptive stream fusion based on feature component rejection for robust multi-stream speech recognition
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Zhang, Jun | - |
dc.contributor.author | Feng, Yizhi | - |
dc.contributor.author | Ning, Gengxin | - |
dc.contributor.author | Ji, Fei | - |
dc.date.accessioned | 2023-12-13T02:00:25Z | - |
dc.date.available | 2023-12-13T02:00:25Z | - |
dc.date.issued | 2015-08 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/116362 | - |
dc.description.abstract | Weighting the stream outputs according to their reliability levels is one of the most common stream fusion methods in the multi-stream automatic speech recognition (MS ASR). However, when a MS ASR system works in noisy environments, there are distortion level differences among not only the data streams, but also the feature components inside a stream. In this paper, we first propose a feature component rejection approach that can provide the similar function as the missing data techniques while is much easier to be applied to different features. Then a new stream fusion method that can make use of the reliability information of both inter-and intra-streams is developed by incorporating the proposed feature component rejection approach into the conventional MS HMM. The proposed stream fusion method shows good noise adaptive ability and achieves similar recognition accuracy as the missing data based stream fusion method for additive noises in the experiments of the Ti digits connected word recognition task. © 2015 IEEE. | - |
dc.format.extent | 5 | - |
dc.language | 영어 | - |
dc.language.iso | ENG | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | Noise adaptive stream fusion based on feature component rejection for robust multi-stream speech recognition | - |
dc.type | Article | - |
dc.publisher.location | 미국 | - |
dc.identifier.doi | 10.1109/ICACI.2015.7184714 | - |
dc.identifier.scopusid | 2-s2.0-84954318805 | - |
dc.identifier.wosid | 000380479600002 | - |
dc.identifier.bibliographicCitation | 2015 Seventh International Conference on Advanced Computational Intelligence (ICACI), pp 279 - 283 | - |
dc.citation.title | 2015 Seventh International Conference on Advanced Computational Intelligence (ICACI) | - |
dc.citation.startPage | 279 | - |
dc.citation.endPage | 283 | - |
dc.type.docType | Conference paper | - |
dc.description.isOpenAccess | N | - |
dc.description.journalRegisteredClass | sci | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Theory & Methods | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.subject.keywordAuthor | Accuracy | - |
dc.subject.keywordAuthor | Hidden Markov models | - |
dc.subject.keywordAuthor | Reliability | - |
dc.identifier.url | https://ieeexplore.ieee.org/document/7184714 | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
55 Hanyangdeahak-ro, Sangnok-gu, Ansan, Gyeonggi-do, 15588, Korea+82-31-400-4269 sweetbrain@hanyang.ac.kr
COPYRIGHT © 2021 HANYANG UNIVERSITY. ALL RIGHTS RESERVED.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.