このアイテムのアクセス数: 173

このアイテムのファイル:
ファイル 記述 サイズフォーマット 
s13634-015-0246-6.pdf2.11 MBAdobe PDF見る/開く
完全メタデータレコード
DCフィールド言語
dc.contributor.authorMimura, Masatoja
dc.contributor.authorSakai, Shinsukeja
dc.contributor.authorKawahara, Tatsuyaja
dc.contributor.alternative三村, 正人ja
dc.date.accessioned2015-11-24T07:34:53Z-
dc.date.available2015-11-24T07:34:53Z-
dc.date.issued2015-07-23-
dc.identifier.issn1687-6172ja
dc.identifier.urihttp://hdl.handle.net/2433/201887-
dc.description.abstractWe propose an approach to reverberant speech recognition adopting deep learning in the front-end as well as b a c k-e n d o f a r e v e r b e r a n t s p e e c h r e c o g n i t i o n s y s t e m, a n d a n o v e l m e t h o d t o i m p r o v e t h e d e r e v e r b e r a t i o n p e r f o r m a n c e of the front-end network using phone-class information. At the front-end, we adopt a deep autoencoder (DAE) for enhancing the speech feature parameters, and speech recognition is performed in the back-end using DNN-HMM acoustic models trained on multi-condition data. The system was evaluated through the ASR task in the Reverb Challenge 2014. The DNN-HMM system trained on the multi-condition training set achieved a conspicuously higher word accuracy compared to the MLLR-adapted GMM-HMM system trained on the same data. Furthermore, feature enhancement with the deep autoencoder contributed to the improvement of recognition accuracy especially in the more adverse conditions. While the mapping between reverberant and clean speech in DAE-based dereverberation is conventionally conducted only with the acoustic information, we presume the mapping is also dependent on the phone information. Therefore, we propose a new scheme (pDAE), which augments a phone-class feature to the standard acoustic features as input. Two types of the phone-class feature are investigated. One is the hard recognition result of monophones, and the other is a soft representation derived from the posterior outputs of monophone DNN. The augmented feature in either type results in a significant improvement (7–8 % relative) from the standard DAE.ja
dc.format.mimetypeapplication/pdfja
dc.language.isoengja
dc.publisherSpringerOpenja
dc.rights© 2015 Mimura et al.ja
dc.rightsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.ja
dc.subjectReverberant speech recognitionja
dc.subjectDeep Neural Networks (DNN)ja
dc.subjectDeep Autoencoder (DAE)ja
dc.titleReverberant speech recognition combining deep neural networks and deep autoencoders augmented with a phone-class featureja
dc.type.niitypeJournal Articleja
dc.identifier.jtitleEURASIP Journal on Advances in Signal Processingja
dc.identifier.volume2015ja
dc.identifier.issue1ja
dc.relation.doi10.1186/s13634-015-0246-6ja
dc.textversionpublisherja
dc.identifier.artnum62ja
出現コレクション:学術雑誌掲載論文等

アイテムの簡略レコードを表示する

Export to RefWorks


出力フォーマット 


このリポジトリに保管されているアイテムはすべて著作権により保護されています。