In this paper, a method for multiple sound source tracking by two microphones is proposed. PSO (Particle Swarm Optimization) is used for searching multiple peaks on MUSIC (MUltiple SIgnal Classification) spectrum. In addition, a difficulty which the tracking fails when a speech returns to a utterance period after a silent period is resolved. Then, a penalty function is added in an evaluation function of PSO, in which a weighted histogram of DOA (Direction-Of-Arrival) estimation results in time-frequency region is used as the penalty. Several experiments were conducted in an actual room to evaluate the effectiveness of the proposed method.
展开▼