This paper describes I2R's submission to the Blizzard Challenge 2010 speech synthesis evaluation. This is our third participation in the challenge. In this paper, we will describe our main approaches to building the required voices. We will introduce the procedure of database processing, the definitions of the acoustic, prosodic and linguistic parameters, the components of cost functions, etc. Finally, we will look at the listening test results. The evaluation results show that our Mandarin system performed well in the evaluation.