Table 1 Research approaches on the task of acoustic surveillance.
From: An Adaptive Framework for Acoustic Monitoring of Potential Hazards
Reference | Atypical sound classes | Model adaptation | Environments | Classifier | Features | Database |
|---|---|---|---|---|---|---|
Proposed approach | Scream, gunshot and explosion | MAP adaptation of GMMs | Metro station, urban and military setting | GMM | MFCC, MPEG-7, CB-TEO, Intonation | Large audio corpora from professional sound effects collections |
Clavel et al. [3] | Fear-type emotions | — | — | GMM | Prosody, audio quality, spectral, cepstral | SAFE corpus |
Valenzise et al. [4] | Scream and gunshot | — | Public square | GMM | Temporal, spectral, cepstral, correlation | Movie soundtracks, internet and people shouts |
Radhakrishnan and Divakaran [5] | Banging and non-neutral speech | — | Elevator | GMM | MFCC | Elevator recordings |
Clavel et al. [6] | Gunshot | — | Public space | GMM | MFCC, spectral moments | CDs for the national French public radio |
Harma et al. [7] | Interesting events in an office | Noise spectrum update | Office | Threshold, clustering | Temporal, spectral | Recordings from an office room |
Rouas et al. [8] | Shout | Adaptive threshold for sound activity detection | Railway | GMM, SVM | Energy, MFCC | Recorded during 4 scenarios |
Vacher et al. [9] | Scream and glass break | — | Apartment | GMM | Wavelet based cepstral coefficients | Laboratory recordings and RCWP |
Atrey et al. [10] | Shout | — | Office corridor | GMM | ZCR, LPC, LPCC, LFCC | Recorded in office corridor |