In this paper, we propose a decision tree based coding mode selection method for the AMR-WB+ audio coder. In order to obtain improved performance with reduced computation, decision tree classifier is adopted with the closed loop mode selection results as the target classification labels. To secure the practical feasibility of this method, the size of the decision tree is controlled by pruning. Through an evaluation on a database covering both speech and music materials, the proposed method is found to increase the mode selection accuracy compared with the open loop mode selection module in the AMR-WB+.