Proper evaluation can efficiently drive the development of text-tospeech (TTS) systems. The assessment is needed to determine how well a system or technique compares to others or how it compares with the previous version of the system. In order to obtain more useful feedback for the development, we do not only evaluate the whole system but also each module of the TTS system separately. Based on the evaluation results, new progress can be achieved on individual modules. Furthermore, the evaluation allows identifying the best techniques in the different modules and comparing different modules for which we have defined a common evaluation specification. To enable the evaluation, the Mandarin TTS system is separated into three modules: text, prosody and acoustic processing modules. Furthermore, common interfaces have been designed for the communication between the TTS modules. The objective and subjective metrics have been proposed for the module-wise evaluation. This paper also presents the evaluation results that our Mandarin TTS modules achieved in the evaluation carried out in the European project on technology and corpora for speech-to-speech translation (TC-STAR). Keywords: Text-to-speech, speech synthesis, SSML, prosodic modeling, evaluation