Automatic categorization of auditory scenes is very useful in various content-based multimedia applications, such as video indexing and context-aware computing. In this paper, an unsupervised approach is proposed to group auditory scenes with similar semantics. In our approach, auditory scenes are described with the key audio effects they contained. In order to exploit the relationships between different audio effects and provide more accurate similarity measure for auditory scene categorization, coclustering is utilized to group the auditory scenes and key audio effects simultaneously. In addition, Bayesian Information Criterion (BIC) is used to automatically select the cluster numbers for both the key effects and the auditory scenes. Evaluation on 272 auditory scenes extracted from 12-hour audio data shows very encouraging results.