DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Younggwan | ko |
dc.contributor.author | Kim, Myungjong | ko |
dc.contributor.author | Goo, Jahyun | ko |
dc.contributor.author | Kim, Hoirin | ko |
dc.date.accessioned | 2018-10-19T00:28:54Z | - |
dc.date.available | 2018-10-19T00:28:54Z | - |
dc.date.created | 2018-09-19 | - |
dc.date.created | 2018-09-19 | - |
dc.date.issued | 2018-11 | - |
dc.identifier.citation | IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, v.26, no.11, pp.2204 - 2214 | - |
dc.identifier.issn | 2329-9290 | - |
dc.identifier.uri | http://hdl.handle.net/10203/245868 | - |
dc.description.abstract | In this paper, we introduce a new feature engineering approach for deep learning-based acoustic modeling, which utilizes input feature contributions. For this purpose, we propose an auxiliary deep neural network (DNN) called a feature contribution network (FCN) whose output layer is composed of sigmoid-based contribution gates. In our framework, the FCN tries to learn element-level discriminative contributions of input features and an acoustic model network (AMN) is trained by gated features generated by element-wise multiplication between contribution gate outputs and input features. In addition, we also propose a regularization method for the FCN, which helps the FCN to activate the minimum number of the gates. The proposed methods were evaluated on the TED-LIUM release 1 corpus. We applied the proposed methods to DNN- and long short-term memory-based AMNs. Experimental results results showed that AMNs with the FCNs consistently improved recognition performance compared with AMN-only frameworks. | - |
dc.language | English | - |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | - |
dc.subject | SPEECH RECOGNITION | - |
dc.subject | NEURAL-NETWORKS | - |
dc.subject | FEATURE-SELECTION | - |
dc.subject | CLASSIFICATION | - |
dc.title | Learning Self-Informed Feature Contribution for Deep Learning-Based Acoustic Modeling | - |
dc.type | Article | - |
dc.identifier.wosid | 000443046300003 | - |
dc.identifier.scopusid | 2-s2.0-85050603905 | - |
dc.type.rims | ART | - |
dc.citation.volume | 26 | - |
dc.citation.issue | 11 | - |
dc.citation.beginningpage | 2204 | - |
dc.citation.endingpage | 2214 | - |
dc.citation.publicationname | IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING | - |
dc.identifier.doi | 10.1109/TASLP.2018.2858923 | - |
dc.contributor.localauthor | Kim, Hoirin | - |
dc.contributor.nonIdAuthor | Kim, Myungjong | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Acoustic modeling | - |
dc.subject.keywordAuthor | deep learning | - |
dc.subject.keywordAuthor | feature contribution network | - |
dc.subject.keywordAuthor | speech recognition | - |
dc.subject.keywordPlus | SPEECH RECOGNITION | - |
dc.subject.keywordPlus | NEURAL-NETWORKS | - |
dc.subject.keywordPlus | FEATURE-SELECTION | - |
dc.subject.keywordPlus | CLASSIFICATION | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.