Attention-based latent features for jointly trained end-to-end automatic speech recognition with modified speech enhancement
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yang, Da-Hee | - |
dc.contributor.author | Chang, Joon-Hyuk | - |
dc.date.accessioned | 2023-10-10T02:30:23Z | - |
dc.date.available | 2023-10-10T02:30:23Z | - |
dc.date.created | 2023-04-06 | - |
dc.date.issued | 2023-03 | - |
dc.identifier.issn | 1319-1578 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/191747 | - |
dc.description.abstract | In this paper, we propose a joint training framework that efficiently combines time-domain speech enhancement (SE) with an end-to-end (E2E) automatic speech recognition (ASR) system utilizing attention-based latent features. Using the latent feature to train E2E ASR implies that various time-domain SE models can be applied for noise-robust ASR and our modified framework is the first approach. We implement a fully E2E scheme pipelined from SE to ASR without domain knowledge and short-time Fourier transform (STFT) consistency constraints by applying a time-domain SE model. Therefore, using the latent feature of time-domain SE as appropriate features for ASR inputs is the main approach in our framework. Furthermore, we apply an attention algorithm to the time-domain SE model to selectively concentrate on certain latent features to achieve the better relevant feature for the task. Detailed experiments are conducted on the hybrid CTC/attention architecture for E2E ASR, and we demonstrate the superiority of our approach compared to baseline ASR systems trained with Mel filter bank coefficients features as input. Compared to the baseline ASR model trained only on clean data, the proposed joint training method achieves 63.6% and 86.8% relative error reductions on the TIMIT and WSJ “matched” test set, respectively. | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | King Saud bin Abdulaziz University | - |
dc.title | Attention-based latent features for jointly trained end-to-end automatic speech recognition with modified speech enhancement | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Chang, Joon-Hyuk | - |
dc.identifier.doi | 10.1016/j.jksuci.2023.02.007 | - |
dc.identifier.scopusid | 2-s2.0-85149206123 | - |
dc.identifier.wosid | 000991156900001 | - |
dc.identifier.bibliographicCitation | Journal of King Saud University - Computer and Information Sciences, v.35, no.3, pp.202 - 210 | - |
dc.relation.isPartOf | Journal of King Saud University - Computer and Information Sciences | - |
dc.citation.title | Journal of King Saud University - Computer and Information Sciences | - |
dc.citation.volume | 35 | - |
dc.citation.number | 3 | - |
dc.citation.startPage | 202 | - |
dc.citation.endPage | 210 | - |
dc.type.rims | ART | - |
dc.type.docType | Article | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | Y | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Information Systems | - |
dc.subject.keywordPlus | OPTIMIZATION | - |
dc.subject.keywordPlus | FRAMEWORK | - |
dc.subject.keywordPlus | NOISE | - |
dc.subject.keywordPlus | CNN | - |
dc.subject.keywordAuthor | Time -domain speech enhancement | - |
dc.subject.keywordAuthor | End -to -end automatic speech recognition | - |
dc.subject.keywordAuthor | Attention -based latent feature | - |
dc.subject.keywordAuthor | Joint training framework | - |
dc.identifier.url | https://www.sciencedirect.com/science/article/pii/S1319157823000368?via%3Dihub | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
222, Wangsimni-ro, Seongdong-gu, Seoul, 04763, Korea+82-2-2220-1365
COPYRIGHT © 2021 HANYANG UNIVERSITY.
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.