000146225 001__ 146225
000146225 005__ 20180913055638.0
000146225 037__ $$aREP_WORK
000146225 245__ $$aEvaluating the Robustness of Privacy-Sensitive Audio Features for Speech Detection in Personal Audio Log Scenarios
000146225 269__ $$a2010
000146225 260__ $$bIdiap$$c2010
000146225 336__ $$aReports
000146225 520__ $$aPersonal audio logs are often recorded in multiple environments. This poses challenges for robust front-end processing, including speech/nonspeech detection (SND). Motivated by this, we investigate the robustness of four different privacy-sensitive features for SND, namely energy, zero crossing rate, spectral flatness, and kurtosis. We study early and late fusion of these features in conjunction with modeling temporal context. These combinations are evaluated in mismatched conditions on a dataset of nearly 450 hours. While both combinations yielded improvements over individual features, generally feature combinations performed better. Comparisons with a state-of-the-art spectral based and a privacy-sensitive feature set are also provided.
000146225 700__ $$0243368$$aParthasarathi, Sree Hari Krishnan$$g177650
000146225 700__ $$0243959$$aMagimai.-Doss, Mathew$$g127186
000146225 700__ $$0243348$$aBourlard, Hervé$$g117014
000146225 700__ $$0241066$$aGatica-Perez, Daniel$$g171600
000146225 8564_ $$uhttp://publications.idiap.ch/downloads/reports/2009/Parthasarathi_Idiap-RR-01-2010.pdf$$zURL
000146225 8564_ $$s456495$$uhttps://infoscience.epfl.ch/record/146225/files/Parthasarathi_Idiap-RR-01-2010.pdf$$zn/a
000146225 909C0 $$0252189$$pLIDIAP$$xU10381
000146225 909CO $$ooai:infoscience.tind.io:146225$$pSTI$$preport
000146225 937__ $$aLIDIAP-REPORT-2010-002
000146225 970__ $$aParthasarathi_Idiap-RR-01-2010/LIDIAP
000146225 973__ $$aEPFL$$sPUBLISHED
000146225 980__ $$aREPORT