The voice of each language usually keeps different syntactic structure. In machine translation, it causes relatively low translation quality. To resolve this problem, an approach is proposed by integrating voice features into hierarchical phrase based (HPB) models. In the proposed method, corpus is firstly classified into three categories from Japanese side: passive voice, potential voice and others. Secondly, passive and potential sentences are classified into several groups according to the characteristics of English to build maximum entropy models for rules. Finally, bilingual voice features are integrated into log linear model for improving translation results and the accuracy of rule selection during the translation of passive and potential sentences. In Japanese to English translation task, large scale experiment shows that the proposed method can not only improve the problem of long distance reordering but also improve translation quality of both passive and potential voice test sets.