This paper describes ILSP and INNOETICS Speech Synthesis System entry for the Blizzard Challenge 2012. A description of the underlying system and techniques used are provided, as well as information about the voice building process and discussion on the obtained evaluation results. Additional focus will be given to new processes or techniques we used this year in comparison to our previous participations, and we will also discuss the results of the new section of the Blizzard Challenge which aims to investigate the abilities of the participating TTS systems to cope with audio books and expressive speech synthesis.