Modular dynamic deep denoising autoencoder for speech enhancement

Deep Denoising Autoencoder (DDAE) is an effective method for noise reduction and speech enhancement. However, a single DDAE with a fixed number of frames for neural network input cannot extract contextual information sufficiently. It has also less generalization in unknown SNRs (signal-to-noise-rati...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2017 7th International Conference on Computer and Knowledge Engineering (ICCKE) s. 254 - 259
Hlavní autori: Safari, Razieh, Ahadi, Seyed Mohammad, Seyedin, Sanaz
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 01.10.2017
Predmet:
Deep Denoising Autoencoder (DDAE)
Improved Context Pattern
Modular Dynamic Deep Denoising Autoencoder (MD-DDAE)
Neural networks
Noise measurement
Noise reduction
Signal to noise ratio
Speech
Speech enhancement
Training
Unseen Signal to Noise Ratio
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Deep Denoising Autoencoder (DDAE) is an effective method for noise reduction and speech enhancement. However, a single DDAE with a fixed number of frames for neural network input cannot extract contextual information sufficiently. It has also less generalization in unknown SNRs (signal-to-noise-ratio) and the enhanced output has some residual noise. In this paper, we use a modular model in which three DDAEs with different window lengths are stacked. Experimental results showes that our proposed architecture, namely modular dynamic deep denoising autoencoder (MD-DDAE) provides superior performance in comparison with the traditional DDAE models in different noisy conditions.
AbstractList Deep Denoising Autoencoder (DDAE) is an effective method for noise reduction and speech enhancement. However, a single DDAE with a fixed number of frames for neural network input cannot extract contextual information sufficiently. It has also less generalization in unknown SNRs (signal-to-noise-ratio) and the enhanced output has some residual noise. In this paper, we use a modular model in which three DDAEs with different window lengths are stacked. Experimental results showes that our proposed architecture, namely modular dynamic deep denoising autoencoder (MD-DDAE) provides superior performance in comparison with the traditional DDAE models in different noisy conditions.
Author Seyedin, Sanaz
Ahadi, Seyed Mohammad
Safari, Razieh
Author_xml – sequence: 1
  givenname: Razieh
  surname: Safari
  fullname: Safari, Razieh
  email: r.safari@aut.ac.ir
  organization: Electr. Eng. Dept., Amirkabir Univ. of Technol., Tehran, Iran
– sequence: 2
  givenname: Seyed Mohammad
  surname: Ahadi
  fullname: Ahadi, Seyed Mohammad
  email: sma@aut.ac.ir
  organization: Electr. Eng. Dept., Amirkabir Univ. of Technol., Tehran, Iran
– sequence: 3
  givenname: Sanaz
  surname: Seyedin
  fullname: Seyedin, Sanaz
  email: sseyedin@aut.ac.ir
  organization: Electr. Eng. Dept., Amirkabir Univ. of Technol., Tehran, Iran
BookMark eNotj7tuwjAUQF2pHVrKD9DFP5DUr_gxooi2CKou7MiPa7BE7MgJA3_fSmU5ZzvSeUGPuWRAaEVJSykx79u-321aRqhqNZVKa_mAlkZp2nEtiSaCP6P1dwnXi6043LIdkscBYPxDLmlK-YTtdS6QfQlQcSwVTyOAP2PIZ5s9DJDnV_QU7WWC5d0LdPjYHPqvZv_zue3X-yYZMjdRa9Mx7gXjUTDtOicENYpJoM4xahkE7V2I0UspnOVOOhNY8D4AZ7JTfIHe_rMJAI5jTYOtt-P9i_8ClzpIFg
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ICCKE.2017.8167886
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Xplore
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Xplore
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781538608043
1538608049
EndPage 259
ExternalDocumentID 8167886
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i90t-f889523c423f428b5b4419726e1bb21a2ed8cbdffc664ba3b6b9d2dccde326573
IEDL.DBID RIE
IngestDate Thu Jun 29 18:37:36 EDT 2023
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i90t-f889523c423f428b5b4419726e1bb21a2ed8cbdffc664ba3b6b9d2dccde326573
PageCount 6
ParticipantIDs ieee_primary_8167886
PublicationCentury 2000
PublicationDate 2017-Oct.
PublicationDateYYYYMMDD 2017-10-01
PublicationDate_xml – month: 10
  year: 2017
  text: 2017-Oct.
PublicationDecade 2010
PublicationTitle 2017 7th International Conference on Computer and Knowledge Engineering (ICCKE)
PublicationTitleAbbrev ICCKE
PublicationYear 2017
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.6407111
Snippet Deep Denoising Autoencoder (DDAE) is an effective method for noise reduction and speech enhancement. However, a single DDAE with a fixed number of frames for...
SourceID ieee
SourceType Publisher
StartPage 254
SubjectTerms Deep Denoising Autoencoder (DDAE)
Improved Context Pattern
Modular Dynamic Deep Denoising Autoencoder (MD-DDAE)
Neural networks
Noise measurement
Noise reduction
Signal to noise ratio
Speech
Speech enhancement
Training
Unseen Signal to Noise Ratio
Title Modular dynamic deep denoising autoencoder for speech enhancement
URI https://ieeexplore.ieee.org/document/8167886
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07a8MwED6S0KFTW5LSNxo61nnIth5jCQktbUOGDNmCpTuRLHZI7P7-SnaaUujSRQghOKRDuu-k7-4AHpE7yyUmkfZ4NQolrvyRcsMIUyFdwBgZ1tn13-VsppZLPW_B0zEWhohq8hn1Q7f-y8fCVuGpbKBG_mpVog1tKUUTq_UdBzPUg9fx-G0SyFqyf5j4q2JKbTCmZ_8TdQ69n8g7Nj_alAtoUd6F548CA1uUYVM-niHR1jd5sQmePsuqsgj5KJF2zGNQtt8S2TWjfB1UGgT1YDGdLMYv0aH0QbTRwzJySmnvIVqPdZz3D0xqPGrRkgsaGcNHGSdU1qBzVojEZLERRiNHa5E8HEtlfAmdvMjpChg3IqHUW2VlvTOQBkBkBCZWxCJOZILX0A2rX22b5Barw8Jv_h6-hdOwwQ2b7Q465a6iezixn-Vmv3uoNfIF396Ozw
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LawIxEB6sLbSntmjpuzn02PWR3U2yxyKK4gMPHrzJJjOhXnZF1_7-JuvWUuillxBCIEyGZL5JZuYDeEVuDZcYBYnDq4GnuHJHynYCjIW0HmOkWFbXn8jZTC2XybwGb8dcGCIqg8-o5bvlXz7mZu-fytqq665WJU7g1DNnVdla35kwnaQ96vXGfR-uJVvV1F-cKaXJGFz-b7EraP7k3rH50apcQ42yBrxPc_TxogwPBPIMiTauyfK19_VZui9yX5ESacscCmW7DZH5YJR9eKX6hZqwGPQXvWFQkR8E66RTBFapxPmIxqEd6zwEHWsncSK5oK7WvJtyQmU0WmuEiHQaaqET5GgMkgNksQxvoJ7lGd0C41pEFDu7rIxzB2IPibTAyIhQhJGM8A4aXvrV5lDeYlUJfv_38AucDxfTyWoymo0f4MJv9iG27RHqxXZPT3BmPov1bvtcaucLgLWSGA
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2017+7th+International+Conference+on+Computer+and+Knowledge+Engineering+%28ICCKE%29&rft.atitle=Modular+dynamic+deep+denoising+autoencoder+for+speech+enhancement&rft.au=Safari%2C+Razieh&rft.au=Ahadi%2C+Seyed+Mohammad&rft.au=Seyedin%2C+Sanaz&rft.date=2017-10-01&rft.pub=IEEE&rft.spage=254&rft.epage=259&rft_id=info:doi/10.1109%2FICCKE.2017.8167886&rft.externalDocID=8167886