WEKO3
アイテム
[Invited Talk] From Bayes Decision Rule to Sequence-to-Sequence Processing by Neural Networks
https://ipsj.ixsq.nii.ac.jp/records/184870
https://ipsj.ixsq.nii.ac.jp/records/184870fabbd397-f80c-4dc2-974c-7c3e7ff64550
名前 / ファイル | ライセンス | アクション |
---|---|---|
![]() |
Copyright (c) 2017 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG.
|
|
SLP:会員:¥0, DLIB:会員:¥0 |
Item type | SIG Technical Reports(1) | |||||||
---|---|---|---|---|---|---|---|---|
公開日 | 2017-12-14 | |||||||
タイトル | ||||||||
タイトル | [Invited Talk] From Bayes Decision Rule to Sequence-to-Sequence Processing by Neural Networks | |||||||
タイトル | ||||||||
言語 | en | |||||||
タイトル | [Invited Talk] From Bayes Decision Rule to Sequence-to-Sequence Processing by Neural Networks | |||||||
言語 | ||||||||
言語 | eng | |||||||
キーワード | ||||||||
主題Scheme | Other | |||||||
主題 | オーガナイズドセッション: 音声言語情報処理が切り拓く新しい情報社会 | |||||||
資源タイプ | ||||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_18gh | |||||||
資源タイプ | technical report | |||||||
著者所属 | ||||||||
RWTH Aachen University | ||||||||
著者所属(英) | ||||||||
en | ||||||||
RWTH Aachen University | ||||||||
著者名 |
Hermann, Ney
× Hermann, Ney
|
|||||||
著者名(英) |
Hermann, Ney
× Hermann, Ney
|
|||||||
論文抄録 | ||||||||
内容記述タイプ | Other | |||||||
内容記述 | With artificial neural networks (ANNs), we have powerful tools to model complex dependencies in Bayes decision rule and go beyond the dependencies in traditional generative models in speech and language processing. In the spirit of hybrid approaches to speech recognition, we re-visit the traditional Hidden Markov model and combine it with ANNs to capture more complex dependencies in both the emission and the transition models. For machine translation, we present first experimental results and compare them with competing approaches. | |||||||
論文抄録(英) | ||||||||
内容記述タイプ | Other | |||||||
内容記述 | With artificial neural networks (ANNs), we have powerful tools to model complex dependencies in Bayes decision rule and go beyond the dependencies in traditional generative models in speech and language processing. In the spirit of hybrid approaches to speech recognition, we re-visit the traditional Hidden Markov model and combine it with ANNs to capture more complex dependencies in both the emission and the transition models. For machine translation, we present first experimental results and compare them with competing approaches. | |||||||
書誌レコードID | ||||||||
収録物識別子タイプ | NCID | |||||||
収録物識別子 | AN10442647 | |||||||
書誌情報 |
研究報告音声言語情報処理(SLP) 巻 2017-SLP-119, 号 14, p. 1-1, 発行日 2017-12-14 |
|||||||
ISSN | ||||||||
収録物識別子タイプ | ISSN | |||||||
収録物識別子 | 2188-8663 | |||||||
Notice | ||||||||
SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc. | ||||||||
出版者 | ||||||||
言語 | ja | |||||||
出版者 | 情報処理学会 |