Automatic recognition of elephant infrasound calls using formant analysis and Hidden Markov Model

A method is proposed for automatic recognition of elephant infrasound calls. It uses Linear Predictive Coding (LPC) for extracting the formants of the elephant rumble. A novel feature extraction technique is proposed to capture the unique features of elephant rumbles. The features are then processed using a two state Hidden Markov Model (HMM) to automatically detect the presence of elephants. The method is tested on pure elephant rumbles as well as field recorded signals containing environmental noise. The results show that the proposed method can be used to successfully detect the presence of elephants even under noisy conditions.