Performance of 3D speaker localization using a small array of microphones

Automatic speaker localization is an important feature in several applications such as videoconferencing and distance learning, where usually only a small array of microphones is available. This feature allows the camera to focus automatically on the person who is speaking. In this work we compare several adaptive filtering algorithms configured as time delay estimators to locate the talker in 3D space. Simulations showing their performance and tracking capability are provided.