A Review on Sound Source Localization Systems

Sound Source Localization (SSL) systems focus on finding the direction of a sound source. Sound source localization is an essential feature in robots and humanoids. Research is being done for two decades to optimize SSL techniques and enhance their accuracy. Presented in this review we have categorized various proposed SSL techniques into four main types. Out of which one type is SSL that is based on conventional algorithms like Generalized Cross Correlation (GCC), Multiple Signal Classification (MUSIC), Time Difference of Arrival (TDOA), etc. using multiple microphones array configurations. The second type involves techniques based on binaural signal processing using conventional algorithms (GCC, MUSIC, TDOA). SSL techniques that fall under the third and fourth types of categories are developed recently in the last decade with the rise of Convolutional Neural Network (CNN) algorithms. The third type of SSL technique makes use of multiple microphone array configurations using CNN and the fourth type involves the most recently evolved technique based on binaural signal processing using CNN. The different SSL techniques based on multiaural and binaural signals using the conventional algorithm as well as CNN are presented in this review. The review paper provides an overview of SSL systems in terms of the number of microphones used, layouts of microphonic arrays, algorithms to perform SSL and localization in 3D space (azimuth, elevation and distance). From the review we found that out of all SSL techniques, CNN is the emerging and optimized one. By using CNN in SSL systems, the least error rate of 0.1 % was achieved.