Mobile audio services are growing with rising popularity of smart mobile devices using WiFi or cellular networks. A major issue facing mobile audio quality assessment is occasional background noises due to the prospect of sound recording at anytime and anywhere with smart mobile devices. Psychological study reveals that people pay selective attention to their interested sound in complex auditory input. In this paper, we model the mobile audio objective quality assessment based on auditory attention mechanism, with attention based horizontal azimuth parameters and timbre distortion parameters as additional Model Output Variables (MOVs). The results show that the prediction accuracy can be obtained by using such a method.