Open AccessDeep Scattering Spectra with Deep Neural Networks for Acoustic Scene Classification Tasks
Author(s) -
Zhang Pengyuan,
Chen Hangting,
Bai Haichuan,
Yuan Qingsheng
Publication year - 2019
Publication title -
chinese journal of electronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.267
H-Index - 25
eISSN - 2075-5597
pISSN - 1022-4653
DOI - 10.1049/cje.2019.07.006
Subject(s) - mel frequency cepstrum , discriminative model , computer science , feature (linguistics) , pattern recognition (psychology) , artificial intelligence , deep neural networks , artificial neural network , feature extraction , speech recognition , task (project management) , engineering , philosophy , linguistics , systems engineering
As one of the most commonly used features, Mel‐frequency cepstral coefficients (MFCCs) are less discriminative at high frequency. A novel technique, known as Deep scattering spectrum (DSS), addresses this issue and looks to preserve greater details. DSS feature has shown promise both on classification and recognition tasks. In this paper, we extend the use of DSS feature for acoustic scene classification task. Results on Detection and classification of acoustic scenes and events (DCASE) 2016 and 2017 show that DSS provided 4:8% and 17:4% relative improvements in accuracy over MFCC features, within a state‐of‐the‐art time delay neural network framework.