Insect Sound Recognition Based on SBC and HMM

In order to help general technicians to recognize insects conveniently in pests management, this paper proposed a viable scheme to identify insect sounds automatically by using Sub-band based cepstral(SBC) and Hidden Markov Model(HMM). The acoustic signal is preprocessed, segmented into a series of sound samples. SBC is extracted from the sound sample as the feature, and HMMs are trained with given features. The matching for a test sample is completed by finding the best matcher in all HMMs. The method is tested in a database with acoustic samples of 50 different insect sounds. The recognition rate was above 90%. The test results proved the efficiency of the proposed method.