Sound source localization plays a significant role in many microphone array applications, ranging from speech enhancement to automatic video conferencing in a reverberant noisy environment. Localization of a sound source by using an algorithm which estimate the location from which the sound is coming from, aids to address; a long time hands-free communications challenge. The steered response power (SRP) using the phase transform (SRP-PHAT) method is one of the most popular modern localization algorithms. The SRP-based source localizers have been proved robust, however, the methods face high computation complexity, making it unsuitable for real time applications and also fail to locate the sound source in inimical noise and reverberation conditions, especially when the direct paths to the microphones are unavailable. This thesis works on a localization algorithm based on discrimination of crosscorrelation functions (CCFs) [1]. The CCFs are determined using the method called the generalized cross-correlation phase transform (GCC-PHAT). Using cross-correlation functions, sound source location is estimated by one of the two classifiers: Naive-Bayes classifier and Euclidean distance classifier. Simulation results have demonstrated that the localization algorithms [1] provide a good accuracy in reverberant noisy environment.
展开▼
