Modification of widely used feature vectors for real-time acoustic events detection

Besides video surveillance system for monitoring large urban areas also the acoustic events detection system can be used. The acoustic detection system is monitoring potentially dangerous sounds and in case of detection an alarm is produced. We developed our own approach to the acoustic events detection system with modified Viterbi decoder operating over HMM (Hidden Markov Models) especially adapted for long-term monitoring task and our own MFCC (Mel-Frequency Cepstral Coeff.) extraction module. In this paper we evaluate our system on new testing database simulating change of environment SNR (Signal-to-Noise Ratio) and also influence of CMN (Cepstral Mean Subtraction) on the detection accuracy. By this occasion we also introduce new modification to our Viterbi decoder. We implemented feature reduction mechanism to omit configurable number of MFC coefficients of input feature vector from decoding process without retraining the HMM models. Results in this paper describe that reduction of feature vector to only delta and acceleration coefficients are improving detection accuracy of our system. We also show in this paper that no CMN is required in front-end even when acoustic model trained with CMN is used.