magyar

Increasing the robustness of CNN acoustic models using autoregressive moving average spectrogram features and channel dropout

Kovács György and Tóth László and Van Compernolle Dirk and Ganapathy Sriram: Increasing the robustness of CNN acoustic models using autoregressive moving average spectrogram features and channel dropout.
PATTERN RECOGNITION LETTERS, 100. pp. 44-50. ISSN 0167-8655 (2017)

[thumbnail of Channel_Dropout_Draft.pdf] Text
Channel_Dropout_Draft.pdf - Accepted Version
Restricted access: Logged in users only
Download (239kB)
Creators:
Kovács György MTMT
Tóth László MTMT
Van Compernolle Dirk
Ganapathy Sriram
Item Type: Journal Article
Journal or Publication Title: PATTERN RECOGNITION LETTERS
Date: 2017
Volume: 100
Page Range: pp. 44-50
ISSN: 0167-8655
Faculty/Unit: Faculty of Science and Informatics
Institution: University of Szeged (2000-)
Language: English
MTMT rekordazonosító: 3279835
DOI azonosító: https://doi.org/10.1016/j.patrec.2017.09.023
Date Deposited: 2018. Feb. 14. 10:29
Last Modified: 2020. Mar. 03. 13:57
URI: http://publicatio.bibl.u-szeged.hu/id/eprint/12962
Web of Science® Times Cited: 29 View citing articles in Web of Science®

Actions (login required)

View Item View Item
Download Statistics

Downloads

Downloads per month over past year

Technical information: publicatio AT ek.szte.hu
SZTE Repository of Publications is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.